A comparative evaluation of the impact of nurse specialists on clinical outcomes in patients with prostate cancer

Transcription

1 Florence Nightingale School of Nursing & Midwifery at King s College London A comparative evaluation of the impact of nurse specialists on clinical outcomes in patients with prostate cancer Briefing Paper Alison Richardson Veronica Tuffrey Lorraine Fincham Sara Faithfull Vincent Khoo Jennifer Wilson Barnett Emma Ream August 2008

2 The Prostate Cancer Charity commissioned this evaluation. Members of the evaluation team: Principal Investigator Professor Alison Richardson, Professor of Cancer and Palliative Nursing Care, King s College London. Dr Emma Ream, Senior Lecturer, King s College London. Professor Sara Faithful, Professor of Cancer Nursing Practice, University of Surrey. Dr Vincent Khoo, Consultant and Honorary Senior Lecturer in Clinical Oncology, Royal Marsden Hospital NHS Foundation Trust and Institute of Cancer Research. Lorraine Fincham, formally Research Associate, King s College London. Dr Veronica Tuffrey, Senior Lecturer, University of Westminster. Professor Dame Jenifer Wilson Barnett, Emeritus Professor, King s College London. For further information about this project please contact the Principal investigator Professor Alison Richardson at the Florence Nightingale School of Nursing & Midwifery, King s College London, 5th Floor, Waterloo Bridge Wing, Franklin Wilkins Building, 150 Stamford Street, London, SE1 9NN Telephone Fax alison.richardson@kcl.ac.uk The research on which this report is based was commissioned by The Prostate Cancer Charity. We would like to express our gratitude to all the teams who contributed to this study, particularly the help extended to us in terms of patient recruitment and willingness to take part in the research. Furthermore, we would like to acknowledge the willingness of The Prostate Cancer Charity Nurse Specialists, Urology Oncology Nurse Specialists and the men with prostate cancer who contributed to the research. The views expressed here are solely those of the authors and are not necessarily representative of the views of The Prostate Cancer Charity King s College London

3 Contents Summary 1 Introduction and Background 2 The Prostate Cancer Charity Specialist Nurse Programme 2 Method 3 Findings 5 Discussion 9 Conclusion 10 References 11

4 Summary Title: A comparative evaluation of the impact of nurse specialists on clinical outcomes in patients with prostate cancer Aim The aim of the study was to compare patient outcomes in those cared for by prostate cancer nurse specialists as compared to those who saw urology oncology nurse specialists. Background The contribution of specialist nurses to improving the care of patients with cancer is well acknowledged. Over the last decade in the United Kingdom there has been a steep rise in the number of nurses dedicated to meeting the needs of patients with specific types of cancer. Patients with prostate cancer have a less satisfactory experience of care than those with other types of cancer. One reason might be the lack of nurses specifically assigned to meet their needs. More information is needed on how the presence of site-specific nurse specialists within cancer multi disciplinary teams influences patient outcomes. Method A quasi experiment was conducted utilising a post-test only design with non equivalent groups. Comparison was made between patients at 2 sites with prostate cancer nurse specialists (Intervention) with patients at 2 matched sites who had access to uro-oncology nurse specialists (Usual Care). Six hundred and fifty-four patients with prostate cancer were surveyed, and comparison of outcomes achieved through examining patients self-reported supportive care needs, urological symptoms, quality of life and experience of, and satisfaction with, care. Results Five hundred and ten patients returned questionnaires (78% response rate). Overall, there was no significant difference in outcomes between patients cared for by a team with a prostate cancer nurse specialists in post and those with access to a uro-oncology nurse specialist. Seventy one (38%) in the Intervention group saw a prostate cancer nurse specialist in the previous six months, whereas only 66 (20%) of the Usual Care group over the same period saw a uro-oncology nurse specialist. In patients who saw a nurse specialist there was a slight trend for those seeing a prostate cancer nurse specialist (as opposed to uro-oncology nurse specialist) to express less unmet needs and more satisfaction with aspects of their care. Conclusion Outcomes in men who see prostate cancer nurse specialists, as opposed to uro-oncology nurse specialists did not differ substantially. This suggests that interventions delivered by both groups of nurses achieve broadly similar effects. With increasing pressure on cancer nurse specialists to justify their influence on patient outcomes thought needs to be given on how site-specific specialist cancer nurses are best deployed to improve the quality and effectiveness of care. 1

5 Introduction and Background The contribution of specialist nurses to improving the experience of care for patients with cancer is well acknowledged. Whilst specialist cancer nurses have been a feature of care in Europe for nearly 30 years, they have recently become key architects in initiatives designed to achieve patient-centred care. In England, the central role of nurses in facilitating the development, co-ordination and delivery of care has been articulated in cancer policy documents (Department of Health 2007). A central tenet of this policy has been the establishment of site specific multi disciplinary teams (MDT), including nurses, who are responsible for planning the treatment and care of patients (Department of Health 2000, Expert Advisory Group on Cancer 1995). Successive national guidance documents have recommended that specialist nurses become a core part of site-specific multidisciplinary teams (NHS Executive 1996, NHS Executive 1998, NHS Executive 2001) and this has led to a dramatic rise in their numbers (Hill 2000). The essence of the role revolves around information provision, emotional support, and care co-ordination. Having a nurse talk things through and interpret what doctors have said regarding diagnosis, prognosis and treatment choices can prove crucial to a patient s understanding of their disease and its management. There has been sustained research interest in clinical nurse specialists as a group. McIntosh (2004) reviewed the evidence and identified the main themes to be: the nature of the role (Casteldine 1995), establishment of new services (McKenna et al. 2004, Moore et al. 2006), expected competencies (Dunn et al. 2000, Cattini and Knowles 1999) and responses of patients to clinical interventions (McCorkle 1989, Moore 2002, Koinberg et al. 2002). Interventions by clinical nurse specialists have thus been evaluated in a wide range of different contexts and in diverse patient groups. Despite sustained research activity in this area the evidence base can still be characterised as weak and uneven (McIntosh 2004). Nurse specialists in cancer care have developed patient care in various situations including, early discharge after breast surgery (Wells et al. 2004), managing follow up of patients (Moore 2002) and on treatment review (Campbell 2000, Faithfull et al. 2001). Corner and colleagues (Corner et al. 2003, Skilbeck et al. 2002, Clark et al. 2002) undertook one of the largest evaluations of specialist nurses in cancer care in an attempt to explore the impact of Macmillan specialist nurses on patients with advanced cancer. This study used a mixed method approach that combined quantitative assessment of outcome with qualitative case studies. They found that patients were often pre occupied with their illness and unable to distinguish Macmillan nurse input from other nursing input. Positive outcomes for patients were attributable to the care provided by the Macmillan nurses, but for a small number some negative outcomes were detected. Because of patients deteriorating conditions it proved difficult to identify impact with any confidence. Whilst we can learn from the research that exists it isn t exhaustive. There remains a continual need to refine the case for establishing and sustaining specialist roles in nursing based on health care need, the impact these have on services and the potential benefits and risks to healthcare organisations of such roles. The Prostate Cancer Charity Specialist Nurse Programme The development of specialist nurse posts in cancer care has mainly been predicated on the needs of patient groups, including those with either breast or colorectal cancer. Yet there is evidence that the experience of care for men with prostate cancer is worse than that of patients with other cancers (National Audit Office and Department of Health 2005). Anecdotally, one reason this might be the case is that this group often do not have access to a nurse dedicated to their care, unlike for example women with breast cancer. Studies have consistently demonstrated the potential of nurse led interventions to manage the uncertainty associated with a diagnosis of prostate cancer (Mishel et al. 2002, Bailey et al. 2004), improve quality of life (Penedo et al. 2003), manage treatment and its side effects (Kim et al. 2002, Faithfull et al. 2001) and enhance the quality of the follow up experience (Booker et al. 2004, Helgesen et al. 2000). Concerned about the poor experience of care men with this type of cancer were receiving, the UK based charity The Prostate Cancer Charity funded a pilot programme to introduce a small number of prostate cancer specific nurse specialists into multi disciplinary teams in the UK. The Prostate Cancer Charity specialist nurse programme aimed to improve the experience of men with prostate cancer through provision of a specialist nursing service. The charity commissioned an evaluation to determine impact. There were a number of elements to the evaluation including a large scale survey of patient experience and a comparison of outcome in sites with, and without, a prostate cancer nurse specialist. This paper reports the comparison of patient outcomes between 2 sites where patients had access to a prostate cancer nurse specialist (sites 2

6 with a prostate cancer nurse specialist) and 2 sites with access to a uro-oncology nurse specialist (sites without a prostate cancer nurse specialist ). The latter were considered to represent the typical care patients might receive in the National Health Service (National Institute for Clinical Excellence 2002). Findings from other elements of the study will be reported elsewhere. Method The aim was to compare outcomes of care provided by prostate cancer nurse specialists with those achieved by urooncology nurse specialists. The primary hypothesis tested was that patients cared for by a team with a prostate cancer nurse specialist would experience less unmet need. Secondary outcomes studied included symptoms, quality of life and satisfaction with care received. It was anticipated not all patients cared for by a MDT would see a specialist nurse. A secondary objective therefore involved examining outcomes of patients who saw a nurse specialist as opposed to those who did not. As randomisation was infeasible a quasi experimental design was adopted to meet the study aim. Moreover, the services were already in situ prior to the evaluation being commissioned. Specifically, a post-test only design with non equivalent groups was utilised (Campbell and Stanley 1963). The comparison involved using naturally occurring groups rather than a control group. The prostate cancer nurse specialist service had been established in the two intervention sites in the previous 18 months. These were matched as closely as possible with two comparison sites that had a uro-oncology nurse specialist, on criteria such as size and type of service and geographical context. Whilst quasi experimental designs are acknowledged to lack power when examining causality they do facilitate examination of causality in situations not conducive to experimental controls (Burns and Grove 1993). Careful assessment of potential threats to validity particular to this design is needed when interpreting findings, such as lack of a pre-test, inability to attribute causality and selection bias. Sample A convenience sample of patients was drawn from the 4 participating sites. The total population of patients with prostate cancer who had been diagnosed between nine and 30 months prior to recruitment were approached. This group would be expected to have had some contact with a nurse specialist and be some way through or have completed their treatment. Other eligibility criteria included: being over18 years of age; aware of their diagnosis; able to read and write English and deemed by their clinical team to be physically and emotionally able to take part. Patients who met these criteria were sent a letter of invitation from their Consultant and asked to reply to the research team if they wished to receive a questionnaire. Data were collected between October 2005 and June Measures Comparison of outcomes was assessed through examining patients reported supportive care needs, urological symptoms, quality of life and experience of, and satisfaction with, care. These outcomes were chosen based on the research team s understanding of the goals of specialist nurses work and published accounts of their contribution to care. This was achieved with the following questionnaires: Supportive care needs The Supportive Care Needs Survey (SCNS) (Bonevski et al. 2000, McElduff et al. 2004) was developed as a comprehensive assessment of the multidimensional impact of cancer and provides an indication patients needs for help in meeting their needs. It has demonstrated good reliability and validity in studies with patients from a range of diagnostic groups. It consists of 41 items designed to assess need in relation to 4 domains: physical and daily living, psychological sexuality, patient care and support, and health system and information. There are five possible answers to choose from in relation to each item: 1. Not applicable this item is not a problem for me as a result of having cancer, 2. Satisfied needing help for this, but need satisfied, 3. Low need the item caused little concern or discomfort, and little need for additional help, 4. Moderate need for help the item caused me some concern or discomfort, some need for additional help, 5. High need for help - this item caused me a lot of concern or discomfort, strong need for additional help. The core module was supplemented with a prostate specific module (Steginga et al. 2001) scored in a similar way. 3

7 Prostate specific symptoms The International Prostate Symptom Score (IPSS) (Barry et al. 1992) is widely used to quantify lower urinary tract symptoms in prostate cancer. The scale consists of seven questions regarding urinary symptoms experienced in the previous month: incomplete emptying, frequency, intermittency, urgency, weak stream, straining and nocturia. Men are asked to score responses on a scale ranging from 0 - not at all to 5 almost always. A further question asks them to rate their quality of life on a scale ranging from 0 delighted to 6 terrible. The first seven items are added together to derive a total score. Quality of life The EuroQol EQ-5D and VAS (EuroQolGroup 1990) is a simple, easy to use tool designed for use across different diseases, languages and countries for describing and valuing health-related quality of life. It is well suited to postal surveys (Brooks 1996). The first section asks subjects to rate their own health state concerning: mobility (no problems, some problems, confined to bed), self care (no problems, some problems, unable to wash/dress themselves), usual activities (no problem, some problems, unable to perform usual activities), pain/discomfort (no pain/discomfort, moderate pain/discomfort, extreme pain/discomfort and anxiety/depression (not anxious or depressed, moderately anxious or depressed, extremely anxious or depressed). The second section, the VAS, asks subjects to rate their health where 100 is the best state imaginable and 0 is the worst state imaginable. A score for health state is computed using the regression model provided with the tool. Experience and satisfaction with care An investigator designed questionnaire was developed as there were no pre-existing tools to assess this variable. It drew on previous work (Faithfull et al. 2001, Moore 2002, Richardson et al. 2002) evaluating the contribution of nurses to improving patient outcomes. The questionnaire comprised 24 items reflecting patients views of: organisation of their care, information and advice given to them, personal experiences of care and general satisfaction with care. A 5-point scale captures responses ranging from 1 not satisfied at all to 6 completely satisfied. In order to establish validity, a principal components analysis with Varimax (orthogonal) rotation was undertaken to check that responses loaded onto factors corresponding to the intended groupings. Through excluding responses to five of the forty-one questions, responses mapped onto the factors as expected. The domain Organisation of Care was comprised of three of these excluded questions so findings for this domain are not reported below. Patients were posted the questionnaire booklet. A reminder letter and additional questionnaire were sent to those who did not reply after 3 weeks. Ethical aspects The study was submitted to, and gained approval from, a multi centre research ethics committee. Return of the questionnaire was taken to indicate consent. Patients were encouraged to contact their clinical team for information and advice if completion raised any questions or concerns. Statistical analysis Data from SCNS and the investigator designed tool to measure experience and satisfaction with care were summarised by creating binary variables i.e. whether a need or dissatisfaction was reported or not. For the SCNS, the category Some need included the responses indicating low, moderate or high need following Lintz et al (Lintz 2003), and for the investigator designed tool measuring satisfaction the category Some dissatisfaction included the responses Not at all satisfied, Barely satisfied and Quite satisfied. Data from other questionnaires were continuous. SPSS v (SPSS Inc., Chicago, II., USA) was used for all statistical analyses, applying a cut-off value for statistical significance of p = 0.05 (two-sided). Baseline values for the demographic and treatment characteristics and outcome measures were compared between usual care and intervention groups using chi-square tests or Fisher s exact test for categorical variables, and the Mann-Whitney test for continuous variables since distributions were skewed. 4

8 Findings Patient characteristics 510 patients returned questionnaires, of which 188 were in sites with prostate cancer nurse specialist (Intervention), and 322 in sites with a uro-oncology nurse specialists (Usual Care). Response rate was 78%. 71 (38%) of patients in the Intervention group reported having seen a prostate cancer nurse specialist, while only 66 (20%) of the Usual Care group reported having seen a uro-oncology nurse specialist. Characteristics of patients included in the two groups are summarised in Table 1, after having combined options for responses for questions about age, education, ethnicity and domestic status. All other variables were derived from questions with Yes/No responses. In this and all other tables, sample sizes are provided for each variable due to missing data. The only statistically significant differences between groups were that the intervention group had a higher proportion of patients from an ethnic minority group (9.6% v 2.5%) and had had a radical prostatectomy (37% v. 25%) and a lower proportion who had had radiation therapy to the prostate (27% v. 36%). Table 1: Participant characteristics Characteristic Total N Intervention Usual Care N (%) N (%) Age greater than 75 years (33.0%) 99 (30.7%) White British Ethnicity* (90.4%) 314 (97.5%) Living with partner/married (84.0%) 272 (84.5%) No formal qualifications (48.1%) 142 (44.5%) Diagnosis more than two years ago (50.3%) 146 (45.3%) State of remission improved (44.7%) 129 (42.7%) Ever had radical prostatectomy* (37.2%) 80 (24.8%) Ever had radiation therapy to the (26.7%) 115 (35.8%) prostate Ever had testicles removed ( 0.5%) 5 ( 1.6%) Ever had hormone therapy (47.9%) 169 (52.5%) Ever had radiation therapy to the bones ( 2.1%) 11 ( 3.4%) Ever had other treatment ( 4.3%) 19 ( 5.9%) Ever had watchful waiting (16.5%) 64 (19.9%) * p< 0.05 In the first instance the comparison of outcomes between patients cared for by teams with a prostate cancer nurse specialist and those cared for by a uro-oncology nurse specialist are described. This is followed by a description of the comparisons between these groups, including only the sub groups of patients who reported actually having seen a nurse specialist. Comparison of groups - entire sample Supportive care needs and views of care Table 2A shows the percentage of patients reporting some need or some dissatisfaction with care according to group. No significant differences were found between the two groups on any domain. Under supportive care needs, the domains with the highest proportion of patients reporting some need were Prostate Symptoms (49.7%) and Psychological (44.0%). The domain with the lowest proportion of patients reporting some need was Patient Care and Support (19.2%). There were no differences between the groups in terms of the percentage of patients expressing some dissatisfaction with care 5

9 Prostate symptoms and self-perceived health status Table 3A shows the distribution of patients according to whether they had mild, moderate or severe prostate symptoms and the percentage reporting urinary incontinence as measured by the IPSS in the Intervention and Usual Care groups. No significant differences were found between the two groups. A higher proportion of those in the Intervention group compared to the Usual Care group fell into the mild symptom category. The proportions of patients falling into the severe category were similar at around 10%. Table 3A also shows the mean EQ 5D and VAS scores indicating Quality of Life and self -valuation of health. For both these variables there were no differences between the Intervention and Usual Care groups. Comparison of groups - sub-sample who reported seeing a nurse specialist Supportive care needs and views of care Table 2B shows the percentage of patients that reported some need or some dissatisfaction, including only the subsample that reported having seen a nurse specialist in the last six months. Generally, the proportions of patients reporting unmet needs for supportive care were higher in Table 2B compared to Table 2A. This indicated that those who saw a nurse specialist tended to express more unmet need than those not seeing a nurse specialist. Moreover, a lower proportion of those in the Intervention group who saw a nurse specialist reported unmet needs compared to those in the Usual Care group who saw a uro-oncology nurse specialist but this did not reach statistical significance. This trend was consistent across all domains except Sexuality. Figure 1 illustrates this trend for the Patient Care and Support domain 6

10 Table 2: Frequency of reporting Some need or Some dissatisfaction by domain and group, for supportive care needs and views of care A Complete sample Domain Total N Intervention Usual Care Supportive care needs N (%) N (%) Physical and daily living (26.2%) 104 (32.6%) Psychological (44.9%) 139 (43.4%) Patient care and support (20.9%) 58 (18.2%) Health system and information (31.6%) 97 (30.2%) Sexuality (35.8%) 99 (31.0%) Prostate Symptoms (49.2%) 161 (50.0%) Views of care Information and advice (64.9%) 179 (62.4%) Personal experience of care (28.1%) 78 (25.0%) Satisfaction with care in general (64.0%) 162 (55.7%) B Including only those who reported having seen a Nurse Specialist Domain Total N Intervention Usual Care Supportive care needs N (%) N (%) Physical and daily living (32.4%) 27(40.9%) Psychological (50.7%) 36 (54.5%) Patient care and support (23.9%) 19 (29.2%) Health system and information (29.6%) 24 (36.4%) Sexuality (40.8%) 26 (40.0%) Prostate Symptoms (54.3%) 40 (60.6%) Views of care Information and advice (59.1%) 36 (62.1%) Personal experience of care (26.8%) 15 (23.4%) Satisfaction with care in general (66.7%) 25 (44.6%) Satisfaction with nurse (43.8%) 34 (58.6%) * p< 0.05 Figure 1 Percentage reporting unmet needs for domain Patient Care and Support on the Supportive Care Needs Survey, by entire sample and in those who reported having seen a Nurse Specialist Intervention Usual care Complete groups Only those who visited specialist nurse 7

11 Table 3 Findings for Prostate Specific Symptoms and Quality of Life by group. A Complete sample International Prostate Symptom score Intervention (n = 186) Intervention (n = 186) N (%) N (%) Mild 109 (58.6%) 160 (50.6%) Moderate 59 (31.7%) 126 (39.9%) Severe 18 (9.7%) 48 (9.5%) Urinary incontinence Intervention (n = 182) Usual Care (n = 302) N (%) N (%) 31 (17.0%) 55 (18.2%) EuroQuol 5D scores Intervention Usual Care N Mean (SD) N Mean (SD) Quality of life and health status score (0.23) (0.19) Scale of valuing your health today (17) (16) B Including only those who reported having seen a Nurse Specialist International Prostate Symptom score Intervention (n = 186) Intervention (n = 186) Intervention (n = 70) Usual Care (n = 65) Mild 41 (58.6%) 31 (47.7%) Moderate 21 (30.0%) 23 (35.4%) Severe 8 (11.4%) 11 (16.9%) Urinary incontinence Intervention (n = 182) Usual Care (n = 302) N (%) N (%) 13 (18.8%) 18 (31.6) EuroQuol 5D scores Intervention Usual Care N Mean (SD) N Mean (SD) Quality of life and health status score (0.28) (0.22) Scale of valuing your health today (21) (15) Under Views of Care on the domain Satisfaction with Care in General, a significantly higher proportion of the Intervention group reported some dissatisfaction compared to the Usual Care group. However, given the multiple statistical tests being applied, this difference could be attributed to chance. In the final domain Satisfaction with Nurse, in which questions related uniquely to the nurse specialist, the trend was noteworthy. Here, a smaller proportion of those who had visited the prostate cancer nurse specialist reported any dissatisfaction compared to those who visited the uro-oncology nurse specialist (p = 0.11). Prostate symptoms and self-perceived health status Table 3B gives the distribution of patients according to the severity of their symptoms and percentage reporting urinary incontinence for the sub-sample who reported having seen a nurse specialist in the last six months. Although the differences between treatment groups were not statistically significant, the trend was for those who saw a prostate cancer nurse specialist to have less severe symptoms. A lower proportion in the prostate cancer nurse specialist group also reported difficulties with urinary incontinence compared to those in the Usual Care group who saw a urooncology nurse specialist. There was no difference in quality of life scores and self-evaluation of health between those who saw a prostate cancer nurse specialist compared to those who saw a uro-oncology nurse specialist in the subsample who reported having seen a nurse (see Table 3B). 8

12 Discussion This quasi experimental study evaluated the impact of prostate cancer nurse specialist s on patients supportive care needs, symptoms, quality of life and experiences of, and satisfaction with, care received compared with Usual Care. There has been little prior work addressing the impact of site specific specialist cancer nurses, rather previous investigators have investigated the impact of specific nursing interventions in different patient groups, including those with prostate cancer. Overall, there was no significant difference in outcomes between patients cared for by a team with a prostate cancer nurse specialist in post and those without. It had been anticipated by the charity developing the programme that introduction of a prostate cancer nurse specialist would bring about change in outcome for patients under the care of a team with a prostate cancer nurse specialist in situ regardless of whether they saw a nurse. It was perhaps ambitious to think that introduction of a single nurse could achieve this magnitude of change in the timeframe involved, and therefore not surprising no differences were found. Moreover, care provided to those in the Usual Care group was delivered by uro-oncology nurse specialists established in their roles for some years and had amassed a considerable amount of clinical experience with the client group. This was not the case with prostate cancer nurse specialist s who were new to their specialist role and had varied experience and were to some extent training on the job. In hindsight a comparison of prostate cancer nurse specialist with uro-oncology nurse specialist was a potentially flawed comparison but as nurse specialist services are now becoming widely established in the UK health service and this meant a naturally occurring and meaningful control are difficult to find. A large proportion of patients in both the Intervention and Usual Care group reported not having seen a nurse specialist. This is not surprising considering recent data describing variation in clinical nurse specialist provision by tumour type that demonstrates significant under provision of nurse specialists for urological cancers (Prostate Cancer Charter for Action 2007). In those patients who saw a nurse specialist in the previous six months there was a suggestion that those seeing a prostate cancer nurse specialist had less unmet need and were less dissatisfied with aspects of their care. On average nurse specialists of either sort had only seen a third to one quarter of patients in the previous six months. Furthermore, the proportion of patients who had seen a nurse specialist in the Usual Care group was much less than those seeing a nurse specialist in the Intervention group. Uro-oncology nurse specialists don t care exclusively for patients with prostate cancer and would have insufficient time to see every patient even if they wanted to. It is not clear how nurse specialists decide who to offer interventions to and anecdotally it appears few of them pursue any form of structured assessment of patient need in order to determine the type of nursing care that might prove beneficial. The introduction of systematic assessment could improve the targeting of a specialist resource at those with most need (Richardson et al. 2007). The fact that nurse specialists in both groups tended to see those with a greater degree of unmet needs suggests they are, albeit in an informal manner, choosing to direct their services at those with more unmet needs. Symptoms associated with prostate cancer and its treatment can be enduring and perhaps it is therefore not surprising that those seeing a nurse continued to report unmet needs. Symptoms are unlikely to disappear or be completely resolved and because of this it could be argued care should be focussed on supporting patients to self-manage their symptoms to optimise living (Foster et al. 2007). Evaluating the impact of specialist nursing roles has been acknowledged as fraught with difficulty (Wilson-Barnett and Beech 1994, McIntosh 2004, Corner 2002). On this occasion there was no opportunity to conduct a randomised trial to compare patient outcomes from services with a prostate cancer nurse specialist in situ as oppose to usual care, nor collect baseline data prior to introducing the intervention. Quasi experimental studies are limited by the extent to which findings can be used to infer causality. This was further impeded by use of a post test only design with non equivalent groups. Interpretation should be undertaken extremely cautiously as there are considerable threats to validity. These include the possibility of selection bias and no pre test against which to assess the impact of the intervention. This creates difficulty in attributing any change in outcome to the intervention. Whilst the sample size overall was reasonable for the comparison between patients cared for by a service with a prostate cancer nurse specialist as oppose to a uro-oncology nurse specialists, the sub group who reported seeing a nurse specialist of either kind in the previous six months was very small. As a result these data should be treated very cautiously. Outcomes were selected based on the research team s understanding of the goals of the prostate cancer nurse specialist programme. Where possible validated instruments were used, but it was necessary to develop a study specific instrument to address experience of, and satisfaction with, care received. This instrument has not been tested for validity or reliability, 9

13 but factor analysis confirmed its inherent structure. It shows promise and should be subject to further development as there are a lack of sensitive outcome measures to determine the measurable effects of care that can be attributed to nurses (Given and Sherwood 2005). It is possible that outcomes selected for investigation did not match closely enough those hypothesised to be affected by the intervention, especially as this was a new role development. Data collected as part of another element of the evaluation (Ream et al. in preparation) confirmed the main emphasis for the prostate cancer nurse specialist was on delivery of information and support and care co-ordination. Differences were however also observed in the degree to which nurse specialist instigated particular interventions e.g. for incontinence and erectile dysfunction or worked with different types of patients e.g. those receiving hormone therapy or in the follow up phase of their illness. This may provide an explanation for why so few differences were observed in relation to the outcomes selected. The intervention was not well defined and was to some degree disparate. Hence, the ability to detect an effect was compromised. Furthermore, the core of any intervention by the two groups of nurse specialist support and information alongside co-ordination of care - was potentially similar across the two groups. This raises the issue of the nature of intervention that nurse specialists might offer. Currently, they work in diverse ways (Clark et al. 2002). Thought needs to be given to how nurse specialists might be best used in future to improve the quality and effectiveness of care (Corner 2003). Hobbs and Murray (Hobbs and Murray 1999) observed on the basis of the evidence that specialist nurses are not likely to have a positive impact on care outcomes if their role is essentially limited to co-ordination, especially if such services themselves are of variable quality. But when nurses have a well defined role in delivering clinical care, offering additional and specialised care such as monitoring treatment, managing follow up or specific patient education and support, they seem to be effective (Booker et al. 2004, Faithfull et al. 2001). A recent review of nursing interventions to improve the health of men with prostate cancer undergoing radiotherapy revealed strong evidence to support nurse-led follow up aimed at helping patients to manage side effects through the provision of information (Tarnhuvud et al. 2007). All this suggests that more impact might be derived if NS concentrated on delivering particular interventions at specific point in the care pathway. Continued thought needs to be given to how nursing roles, many of which are new, might be best used to improve the quality and effectiveness of care (Corner 2003). Conclusion In conclusion, findings suggest patient outcomes in men who saw a prostate cancer nurse specialist as opposed to uro-oncology nurse specialist did not differ substantially. The care delivered by both sets of nurses achieved similar effects. Alternative explanations are however equally plausible. The observation that nurse specialist in this study were seeing patients with the most unmet need, and those who saw a prostate cancer nurse specialist as opposed to a uro-oncology nurse specialist were more satisfied with certain aspects of the care experience and experienced less unmet need is encouraging. Further well designed and sufficiently powered studies using sensitive outcome measures are needed to determine if trends detected were a true effect or merely occurred by chance. Whilst the contribution site specific nurse specialists make to patient care is well rehearsed in the literature we should re examine how they might bring about most impact on the patient experience of care and subsequent outcomes. Questions such as How do specialist nurses benefit patients? and In what ways are they effective? need to be answered (Corner 2002). If randomised studies do not prove to be feasible those instigating new nursing innovations need to consider evaluation at an early stage of service development and make every effort to collect baseline data in order to assess the degree of change realised. 10

14 References Bailey, D., Mishel, M., Belyea, M., Stewart, J. and Mohler, J. (2004) Uncertainty intervention for watchful waiting in prostate cancer. Cancer Nursing, 27(5), Barry, M., Fowler, F. J., O Leary, M., Bruskewitz, R., Holtgrewe, H., Mebust, W., Cockett, A. and on behalf of the Measurement Committee of the American Urological Association (1992) The American Urological Association symptom index for benign prostatic hyperplasia: the measurement committee of the American Urological Association. Journal of Urology, 148(5), Bonevski, B., Sanson-Fisher, R., Girgis, A., Burton, L., Cook, P. and Boyes, A. (2000) Evaluation of an instrument to assess the needs of patients with cancer. Cancer, 88(1), Booker, J., Eardley, A., Cowan, R., Logue, J., Wylie, J. and Caress, A. (2004) Telephone first post-intervention follow-up for men who have had radical radiotherapy to the prostate: evaluation of a novel service delivery approach. European Journal of Cancer Care, 8, Brooks, R. w. t. E. G. (1996) EuroQol: the current state of play. Health Policy, 37, Burns, B. and Grove, K. (1993) The practice of nursing research, W.B. Saunders, Philadelphia. Campbell, D. and Stanley, J. (1963) Experimental and quasi experimental designs for research, Rand McNally, Chicago. Campbell, J., German, L., Lane, C., Dodwell, D. (2000) Radiotherapy outpatient review: A nurse-led clinic. Clinical Oncology, 12, Casteldine, G. (1995) Defining specialist nursing. British Journal of Nursing, 4, Cattini, P. and Knowles, V. (1999) Core competencies for clinical nurse specialists: a usable framework. Journal of Clinical Nursing, 8, Clark, D., Seymour, J., Douglas, H., Bath, P., Beech, N., Corner, J., Halliday, D., Hughes, P., Haviland, J., Normand, C., Marples, R., Skilbeck, J. and Webb, T. (2002) Clinical nurse specialists in palliative care. Part 2. Explaining diversity in the organisation and costs of Macmillan nursing services. Palliative Medicine, 16, Corner, J. (2002) Evaluating the work of clinical nurse specialists in palliative care. Palliative Medicine, 16, Corner, J. (2003) The role of nurse-led care in cancer management The Lancet, 4, Corner, J., Halliday, D., Haviland, J., Douglas, H., Bath, P., Clark, D., Normand, C., Beech, N., Hughes, P., Marples, R., Seymour, J., Skilbeck, J. and Webb, T. (2003) Exploring nursing outcomes for patients with advanced cancer following intervention by Macmillan specialist palliative care nurses. Journal of Advanced Nursing, 41(6), Department of Health (2000) The NHS Cancer Plan. A plan for investment. A plan for reform, Department of Health, London. Department of Health (2007) Cancer Reform Strategy, Department of Health, London. Dunn, S., Lawson, D., Robertson, S., Underwood, M., Clark, M., Valentine, T., Walker, N., Wilson-Row, C., Crowder, K. and Herewane, D. (2000) The development of competency standards for specialist critical care nurses. Journal of Advanced Nursing, 31(2), EuroQolGroup (1990) A new facility for the measurement of health-related quality of life. Health Policy, 16, Expert Advisory Group on Cancer (1995) A policy framework for commissioning cancer services, Department of Health, London. Faithfull, S., Corner, J., Meyer, L., Huddart, R. and Dearnaley, D. (2001) Evaluation of nurse-led follow up for patients undergoing pelvic radiotherapy. British Journal of Cancer, 85(12), Foster, C., Brown, B. and Killen, M. (2007) The NCRI Cancer Experiences Collaborative: defining self-management. European Journal of Oncology Nursing, 11, Given, B. and Sherwood, P. (2005) Nursing-sensitive patient outcomes - a white paper. Oncology Nursing Forum, 11

15 32(4), Helgesen, F., Andersson, S., Gustafsson, O., Varenhorst, E., Goben, B., Carnock, S., Sehlstedt, L., Calsson, P., Holmberg, L. and Johansson, J. (2000) Follow-up of prostate cancer patients by on-demand contacts with a specialist nurse. Scandinavian Journal of Urology and Nephrology, 34, Hill, A. (2000) The impact of expanding the numbers of clinical nurse specialists in cancer care: a United Kingdom case study. European Journal of Oncology Nursing, 4(4). Hobbs, R. and Murray, G. (1999) Specialist liaison nurses: Evidence for their effectiveness. British Medical Journal, 318, Kim, Y., Roscoe, J. and Morrow, G. (2002) The effects of information and negative affect on severity of side effects from radiation therapy for prostate cancer. Supportive Care in Cancer, 10(5), Koinberg, I., Holmberg, L. and Fridland, B. (2002) Breast cancer patients satisfaction with a spontaneous check-up visit to a specialist nurse. Scandinavian Journal of Caring Science, 16, Lintz, K.,, Moynihan, C., Steginga, Suzanne, Norman, A., Eeles, R., Huddart, R., Dearnaley, D., Watson, M. (2003) Prostate cancer patients support and psychological care needs: survey from a non-surgical oncology clinic. Psycho- Oncology, 12, McCorkle, R., Benoliel, J., Donaldson, G., Georgiadou, Fortini, Moinpour, C., Goodell, B. (1989) A randomized clinical trial of home nursing care of lung cancer patients. Cancer, 64, McElduff, P., Boyes, A., Zuccas, A. and Girgis, A. (2004) Supportive Care Needs Survey: a guide to administration, scoring and analysis. Centre for Health Research and Psycho-oncology, University of Newcastle, Australia. McIntosh, J. (2004) Evaluating the evidence for further investment in the clinical nurse specialist and nurse consultant role. Key note paper at New Nursing Roles: Deciding the future for Scotland. Role Consensus Conference November Royal College of Physicians Edinburgh. Scottish Executive, Edinburgh. McKenna, H., McCann, S., McCaughan, E. and Keeney, S. (2004) The role of an outreach oncology nurse practitioner: A case study evaluation. European Journal of Oncology Nursing, 8, Mishel, M., Belyea, M., Germino, B., Stewart, J., Bailey, D., Robertson, C. and Mohler, J. (2002) Helping patients with localised prostate carcinoma manage uncertainty and treatment side effects. Cancer, 94(6), Moore, S., Corner, J., Haviland, J., Wells, M., Salmon, E., Normand, C., Brada, M., O Brien, M., Smith, I. (2002) Nurse led follow up and conventional medical follow up in management of patients with lung cancer: randomised trial. BMJ, 325, Moore, S., Wells, M., Plant, H., Fuller, F., Wright, M. and Corner, J. (2006) Nurse specialist follow-up in lung cancer: the experience of developing and delivering a new model of care. European Journal of Oncology Nursing, 10, National Audit Office and Department of Health (2005) Tackling cancer: Improving the patient journey, The Stationary Office, London. National Institute for Clinical Excellence (2002) Guidance on cancer services. Improving outcomes in urological cancers. The manual, National Institute for Clinical Excellence, London. NHS Executive (1996) Improving outcomes in breast cancer. The manual, NHS Executive, London. NHS Executive (1998) Improving outcomes in lung cancer. The manual, NHS Executive, London. NHS Executive (2001) Improving outcomes in upper gastro-intestinal cancers. The manual, NHS Executive, London. Penedo, F., Dahn, J., Molton, I., Gonzalez, J., Kisinger, D., Roos, B., Carver, C., Schneiderman, N. and Antoni, M. (2003) Cognitive behavioural stress management improves stress-management skills and quality of life in men recovering from treatment of prostate cancer. Cancer, 100(11), Prostate Cancer Charter for Action (2007) Because Men Matter. The Case for Clinical Nurse Specialists in Prostate Cancer. Prostate Cancer Charter for Action, London. 12

16 Ream, E., Wilson-Barnett, J., Faithfull, S., Fincham, L., Khoo, V. and Richardson, A. (in preparation) Clinical Nurse Specialists in prostate cancer: a description of the process and outcome of care. Richardson, A., Halliday, D. and Wilson-Barnett, J. (2002) Evaluating the impact of introducing lung and colorectal nurse specialists in the South East London Cancer Network, King s College London, London. Richardson, A., Tebbit, P., Brown, V., Sitzia, J. and on behalf of the Cancer Action Team (2007) Assessment of Supportive and Palliative Care Needs for Adults with Cancer, King s College London, London. Skilbeck, J., Corner, J., Bath, P., Beech, N., Clark, D., Hughes, P., Douglas, H., Halliday, D., Haviland, J., Marples, R., Normand, C., Seymour, J. and Webb, T. (2002) Clinical nurse specialists in palliative care. Part 1. A description of the Macmillan Nurse caseload. Palliative Medicine, 16, Steginga, S., Occhipinti, S., Dunn, J., Gardiner, R., Heathcote, P. and Yaxley, J. (2001) The supportive care needs of men with prostate cancer. Psycho-oncology, 10, Tarnhuvud, M., Wandel, C. and Willman, A. (2007) Nursing interventions to improve the health of men with prostate cancer undergoing radiotherapy: a review. European Journal of Oncology Nursing, 11, Wells, M., Harrow, A., Donnan, P., Davey, P., Devereux, S., Little, G., McKenna, E., Wood, R., Chen, R. and Thompson, A. (2004) Patient, carer and health service outcomes of nurse-led early discharge after breast cancer surgery: a randomised controlled trial. British Journal of Cancer, 91, Wilson-Barnett, J. and Beech, S. (1994) Evaluating the clinical nurse specialist. A review. International Journal of Nursing Studies, 31(6),