Showing papers in "Journal of Health Services Research & Policy in 2004"
873 citations
TL;DR: Generalised linear models are attractive for the regression of cost data because they provide parametric methods of analysis where a variety of non-normal distributions can be specified and the way covariates act can be altered.
Abstract: Objective: Choosing an appropriate method for regression analyses of cost data is problematic because it must focus on population means while taking into account the typically skewed distribution o...
362 citations
TL;DR: The wider literature suggested that factors promoting success include: introducing ‘treatments’ of proven efficacy; appropriate staff education and training; removal of unhelpful boundary demarcations between staff or service sectors; appropriate pay and reward systems; and good strategic planning and human resource management.
Abstract: OBJECTIVE: Changing workforce skill-mix is one strategy for improving the effectiveness and efficiency of health care. Our aim was to summarise available research into the success or failure of skill-mix change in achieving planned outcomes. METHODS: A systematic search for existing reviews of research into skill-mix was conducted. Databases searched included: MEDLINE, CINAHL, PsychINFO, Cochrane Library, HMIC, Centre for Reviews and Dissemination, and Department of Health Research Findings Register. Search terms included keywords defining the type of publication, clinical area, type of health personnel and the focus of the article (role change, skill-mix, etc.). English language publications from 1990 onwards were included. Two reviewers independently identified relevant publications, graded the quality of reviews and extracted findings. In addition, the wider literature was scanned to identify which factors were associated with the success or failure of skill-mix change. RESULTS: A total of 9064 publications were identified, of which 24 met our inclusion criteria. There was a dearth of research, particularly for role changes involving workers other than doctors or nurses. Cost-effectiveness was generally not evaluated, nor was the wider impact of change on health care systems. The wider literature suggested that factors promoting success include: introducing 'treatments' of proven efficacy; appropriate staff education and training; removal of unhelpful boundary demarcations between staff or service sectors; appropriate pay and reward systems; and good strategic planning and human resource management. Unintended consequences sometimes occurred in respect of: staff morale and workload; coordination of care; continuity of care; and cost. CONCLUSIONS: In order to make informed choices, health care planners need good research evidence about the likely consequences of skill-mix change. The findings from existing research need to be made more accessible while the dearth of evidence makes new research necessary.
323 citations
TL;DR: An overview of alternative approaches to modelling in economic evaluation, and to highlight situations where each of the alternative modelling techniques should be employed, to highlight the crucial question to be answered when selecting the approach to modelling.
Abstract: Objectives: To provide an overview of alternative approaches to modelling in economic evaluation, and to highlight situations where each of the alternative modelling techniques should be employed.Methods: A review of the available approaches to modelling in the economic evaluation of health care interventions with a leading discussion of examples of published studies leading to guidance in the selection of an appropriate approach in different circumstances.Results: The main approaches to modelling used in economic evaluations in health care are decision trees, Markov models and individual sampling models. These methods assume independence of individuals within the model. Where interaction between individuals is important, other methods such as discrete-event simulation or system dynamics are preferable.Conclusions: The paper highlights the crucial question to be answered when selecting the approach to modelling: can the individuals being simulated in the model be regarded as independent? This issue is ver...
282 citations
TL;DR: Temporospatial constraints and contrasting notions of 'collaboration' are important factors in shaping the nature of interprofessional collaboration, and policies that promote this activity cannot assume that shared understanding of collaboration exists.
Abstract: Objectives: Interprofessional collaboration is widely advocated in health and social care policies. However, existing research provides a relatively poor understanding of how professionals collaborate or the meanings they attach to their collaborative work. This paper aims to contribute to understanding of this activity by providing an in-depth account of interprofessional collaboration on medical wards in a large teaching hospital.Methods: Ethnographic methods were used, including individual and group interviews with health and social care staff (i.e. doctors, nurses, therapists, social workers; n = 49) and participant observations of ward-based work.Results: The organisation of medical teams, who cared for patients across a number of wards, and the task-oriented nature of medical work, limited opportunities for collaboration with other professionals (e.g. nurses, therapists) who were usually based on a single ward. Consequently, collaboration tended to be task-based, terse and formalistic. Staff employe...
195 citations
TL;DR: Inclusion of standardised survey questions in qualitative interviews can provide an easy and fruitful method to explore research issues and provide triggers to difficult or contested topics.
Abstract: Objectives: Multi-method approaches are increasingly advocated in health services research (HSR). This paper examines the use of standardised self-completion questionnaires and questions, during in-depth interviews, a technique termed 'questerviews'.Methods: 'Questerview' techniques were used in four empirical studies of health perceptions conducted by the authors. The studies included both standardised self-completion questions or questionnaires and in-depth interviews. Respondents were tape-recorded while they completed the standardised questionnaires and were encouraged to discuss their definitions of terms and responses to items in-depth. In all studies, 'questerviews' were fully transcribed and data analysis involved the scrutinising of transcripts to identify emergent themes.Results: Responses to the standardised items led to rich sources of qualitative data. They proved to be useful triggers as respondents discussed their understanding and definitions of terms, often explaining their responses with...
119 citations
TL;DR: Some basic intellectual and institutional features of interdisciplinary research are identified, typical interdisciplinary 'dreams' and corresponding 'nightmares' that researchers might encounter in practice are described, and implications for health research policy are discussed.
Abstract: Interdisciplinarity has become popular in health services research. Advocates suggest that interdisciplinary approaches may produce more accessible, applicable, exciting and realistic knowledge than traditional disciplinary approaches. To date, there has been surprisingly little analysis of the institutional and intellectual demands of interdisciplinarity as a methodology or practice. This paper (1) identifies some basic intellectual and institutional features of interdisciplinary research, (2) describes typical interdisciplinary 'dreams' and corresponding 'nightmares' that researchers might encounter in practice, (3) highlights maladies of interdisciplinary research careers and suggests practical remedies, and (4) discusses implications for health research policy. Individual researchers can avoid pitfalls of interdisciplinarity through strategies that include selective collaboration, cross-training, sustained relationships, good humour, participation in peer review, declaring the place of one's work, and...
77 citations
TL;DR: The findings suggest that citizen participants are highly critical of, and discerning about, their public participation experiences, Yet, they are optimistic and determined to contribute in meaningful ways to future public policy processes.
Abstract: Introduction: Health policy decision-makers are grappling with increasingly complex and ethically controversial decisions at a time when citizens are demanding more involvement in these decision processes.Objectives: To assess and revise a set of guiding principles for the design of public involvement processes generated from a synthesis of public participation design and evaluation frameworks that can be used to inform the design and evaluation of future public participation processes in the health sector.Methods: Six focus groups held in five Canadian provinces comprising citizens with considerable experience of public participation processes.Results and discussion: Our findings suggest that citizen participants are highly critical of, and discerning about, their public participation experiences. Yet, they are optimistic and determined to contribute in meaningful ways to future public policy processes. They are clear about where improvements are needed and give top priority to what information is shared...
74 citations
TL;DR: In examining the perspectives of decision-makers in using evidence to support priority-setting, valuable information was derived which should provide insight for such processes in other jurisdictions.
Abstract: Objectives: Resource scarcity dictates the need for health organisations to set priorities. Although such activity should be based, at least in part, on evidence, there are limited examples in the ...
72 citations
TL;DR: The paper concludes by noting the positive aspects of conducting qualitative research in health economics: the interesting and motivating nature of the research and, particularly, the possibilities for increasing the relevance associated with economic theory in the context of health and health services.
Abstract: As health economists begin to embrace qualitative methodology they inevitably face a number of issues. This paper explores these, distinguishing between those associated with the conflict between quantitative and qualitative methodologies (that has already been faced in a number of other research areas) and those associated with the potential for challenges to the discipline of mainstream economics. The former include both the acceptability of the methods (because of issues such as generalisability and reflexivity) and the acceptability of presentation. The latter appear to be essentially concerned with identity within economics. The paper concludes by noting the positive aspects of conducting qualitative research in health economics: the interesting and motivating nature of the research and, particularly, the possibilities for increasing the relevance associated with economic theory in the context of health and health services.
66 citations
TL;DR: This paper can be safely elected to the health services research ‘hall of fame’, since the ‘tracer’ concept it proposed is now taken for granted as a tool for assessing health care quality.
Abstract: David Kessner and colleagues’ paper can be safely elected to the health services research ‘hall of fame’, since the ‘tracer’ concept it proposed is now taken for granted as a tool for assessing health care quality. The authors proposed that a set of common health problems be chosen as tracers of quality. Medical records for patients with these problems would be sampled and reviewed, and actual care compared with a predetermined minimal standard of good practice. The example the authors used was the care of hypertension in a neighbourhood clinic where they found that 30% of patients’ care did not meet minimal requirements for good care. They suggested that this approach could be used to compare care between providers. The authors proposed that tracers should be frequent, well defined conditions affected by medical care, with agreed appropriate care and with a known epidemiology. They suggested conditions such as otitis media, visual disorders, iron deficiency anaemia, hypertension and cervical cancer be used as tracers. The 30 years since this publication have seen transformation of the tracer idea in response to wider changes in health care systems and health care analysis.
TL;DR: Directions for future research and investment in this area are proposed, including the need for a stronger theoretical base and for longitudinal studies over extended periods of time in order to examine short, medium and long-term outcomes in relation to health care practice.
Abstract: This paper addresses the theme of interprofessional education for health and social care professionals as it affects the development of the workforce. The drivers for change in the UK, typified by the Bristol Royal Infirmary and Victoria Climbie inquiries and the response to these in the form of Department of Health policy, are discussed. The need for rapid development of the evidence base around this subject is evident from literature reviews of the impact of interprofessional education. Directions for future research and investment in this area are proposed, including the need for a stronger theoretical base and for longitudinal studies over extended periods of time in order to examine short, medium and long-term outcomes in relation to health care practice.
TL;DR: The TTO elicited from the patient perspective, as currently practised, should not be used as an input for QALYs or for comparisons of diagnostic groups.
Abstract: Objective: Cost-utility analysis is gaining importance as a tool for setting priorities in health care. The approach requires quality-of-life weights on a scale from 0.00 (corresponding to death) to 1.00 (corresponding to perfect health). Different methods and perspectives of the evaluators tend to give different results. Time trade-off (TTO) is the most commonly used method to elicit quality-of-life weights for quality-adjusted life-years (QALYs). How reliable are the results of this method, when limited to one specific perspective, as input for cost-utility analysis?Method: Systematic literature review of empirical studies in which the TTO is elicited by the respondent on their own behalf.Results: In 56 papers, quality-of-life weights for 102 diagnostic groups were given. Ranking of the diagnostic groups according to their quality-of-life weights had no apparent relation to severity. One specific diagnostic group was assigned quality-of-life weights ranging from 0.39 to 0.84. Altogether, 57% of responde...
TL;DR: A greater understanding of the importance of the engagement process and the selection of appropriate methods for the objectives the authors are trying to achieve is argued, as the authors argue.
Abstract: Patient and public involvement is central to the British government’s modernisation of the National Health Service. A ‘patient-centred service’, it is argued, is necessary to improve health outcomes, provide a better experience of care for patients and reconnect the health service with the communities it serves.1 New structures are being put in place to enable and promote these changes, including patient and public involvement forums in every NHS Trust, and public membership and governors of foundation hospitals, backed by a legal duty to consult the public on all significant changes in services or provision.2 Parallel approaches are developing in Canada, Australia, the USA and in some other European countries. The paper from Julia Abelson and colleagues (pp 205–212) from Canada is a timely reminder of the difficulties of, and opportunities for, public involvement. They find, significantly, that citizens are both conscious and critical of the processes of consultation and involvement, and ‘discerning about their public participation experiences’. Citizen participants are to some extent cynical about the intentions and purposes of public consultations, but are, nonetheless, ‘optimistic and determined to contribute’. Abelson and colleagues’ paper highlights the lack of clarity of purpose of some public engagement processes. Put simply, what are they for? Are they to re-engage citizens in the democratic process, to connect public services with communities, to push up quality and performance, to improve effectiveness, to increase patient satisfaction or to create better health outcomes? All these objectives have been cited as justifications for patient and public involvement. Each of them may in itself be desirable and they are not mutually exclusive or contradictory, but we do need, as the authors argue, a greater understanding of, and concentration on, the importance of the engagement process and the selection of appropriate methods for the objectives we are trying to achieve. A patient survey, for instance, is not a good way of eliciting views on complex issues such as service reconfiguration, but is a good means of gathering information about patients’ experiences. A citizens’ jury may be helpful in getting public engagement in public health policy, but is not appropriate if we are seeking patient involvement in improving booking systems. Patient and public, or citizen, involvement are usually linked in policy and practice. They are not, of course, the same thing, though they can be regarded as two sides of the same coin. We take on different roles when we engage with health services as patients or carers or community representatives or citizens or taxpayers. We ask different things of them in our different roles and the dialogue between us needs to reflect that. Both public and patient involvement have at their heart the aspiration towards better health outcomes; better for individuals because they are informed, self-managing, making choices and using health services effectively; better for communities because the health care system becomes more responsive, more directed to customer service and quality, and more connected to local need. The evidence that involvement can do this is building slowly. A recent review in England looked at 12 studies involving patients and concluded:
TL;DR: Changes occurring in the nature of professional practice suggest a fundamental shift in the social relations of health care and the role of medicine.
Abstract: This paper explores the nature and future of social research on patient-professional interactions. It first sketches the historical background to such research and notes that in the UK and US this was characterised by a focus on the doctor-patient relationship. This research embodied a sceptical view of the power of the medical profession in sustaining and promoting social inequalities, and a critique of 'medical dominance' over other health care professionals and patients. The paper then goes on to outline changes occurring in the nature of professional practice that suggest a fundamental shift in the social relations of health care and the role of medicine. These include a putative loss of public confidence in the medical profession and the authority of science, an increased role of the media in informing patients, and a change in the state's relationship with health care professionals. Finally, the paper outlines some items for a future research agenda, including the need to understand better patient preferences about changes in health care delivery, including a willingness to engage in 'partnership', and the possibilities and barriers to change in professional practice.
TL;DR: The outcomes examined in this study were broader than those used in conventional forms of economic evaluation and this enabled identification and analysis of more diverse sources of value from this programme.
Abstract: OBJECTIVES: To conduct an holistic economic evaluation of an Aboriginal community (Daruk)-controlled midwifery service in Western Sydney. METHODS: The study estimated direct programme costs and downstream savings to the health sector. Measures of antenatal attendance and perinatal outcome were compared between Aboriginal women with and without experiences of the antenatal programme through examination of antenatal clinic records and the New South Wales Midwives Data Collection for the period 1991-1996. Qualitative outcomes were analysed through interviews and focus groups. RESULTS: The net cost to the health sector was estimated to be $A1200 per client. Compared with Aboriginal women who received antenatal care in two conventional settings, there were no statistically significant differences in birthweight and perinatal survival. Gestational age at first visit was lower (17.2 vs 21.2 and 19.9 weeks), mean number of antenatal visits was higher (10.5 vs 5.5 and 9.5) and attendance for routine antenatal tests was better (94% vs 71% and 84%). Aboriginal clients were strongly positive about their experience of the Daruk service in terms of relationship and trust, accessibility, flexibility, provision of clear and appropriate information, continuity of care, empowerment and family-centred care. CONCLUSIONS: Net per client costs to the health service were modest. Quantitative evidence of improved antenatal outcomes was limited, but qualitative evaluation suggested the service was strongly valued. The outcomes examined in this study were broader than those used in conventional forms of economic evaluation and this enabled identification and analysis of more diverse sources of value from this programme.
TL;DR: The range of patient pathways following surgery for colorectal cancer and patients' needs and preferences for follow-up are identified and a list of topics for GPs to discuss with their patients about following-up is suggested.
Abstract: Objectives: To identify the range of patient pathways following surgery for colorectal cancer and explore patients' needs and preferences for follow-up.Methods: A survey of hospitals within the UK ...
TL;DR: Survival was poorer in patients treated in the public hospital system, but the possession of private health insurance was not predictive of better outcomes.
Abstract: ObjectivesThe effects of demographic, locational and socio-economic disadvantage, and the influence of private health care on five-year mortality rates in patients with lung cancer or after breast cancer surgery in Western Australia were examined.MethodsThe Western Australian Record Linkage Project was used to extract all hospital morbidity, cancer and death records of all people with lung or breast cancer in Western Australia from 1982 to 1996. Mortality rate ratios after a diagnosis of lung cancer or breast cancer surgery were estimated using Cox regression. Two sets of analyses were carried out: demographically adjusted from 1982 to 1996; and demographically and disadvantage adjusted from 1992 to 1996.ResultsOverall, 87.7% of lung cancer and 17.8% of breast cancer patients were deceased by five years. Lung and breast cancer patients treated in rural hospitals had higher mortality rates (1992-1996: relative risk (RR) 1.24, 95% confidence interval (CI) 1.07-1.44, and RR 1.20, 95% CI 0.92-1.56, respective...
TL;DR: 'Maximum part-time contract' status is associated with higher absolute activity rates among NHS surgeons, and a multi-level model indicates that consultant surgeons with a 'maximum part- time' contract have significantly higher activity rates than those with a full time contract.
Abstract: Objectives: To explore variation in the National Health Service (NHS) activity rates of consultant surgeons, and examine whether activity is dependent upon the type of contract held and/or bonus pa...
TL;DR: Three main themes emerged from the data: patients recognized the need to balance their consent preferences with time pressures in the clinical encounter when deciding the nature of consent for a study; patients generally regarded the seeking of consent as being an issue of respect for them as individuals.
Abstract: Objective: To explore the consent preferences of patients whose health data are currently being used for research purposes.Methods: Semi-structured interviews were conducted with 17 patients whose ...
TL;DR: The recent ferment surrounding professional self-regulation in medicine and other health professions is explored and there is a need to explore the changing relationship between the state and professions and implications, not only for the professions but for health care more broadly.
Abstract: This article explores the recent ferment surrounding professional self-regulation in medicine and other health professions. It reviews the academic literature and sets out an agenda for research. T...
TL;DR: There was no evidence that the clinical background of nurses, their length of experience in NHS Direct, the range of their experience, or their gender affected triages, and the appropriateness of triage decisions still needs to be evaluated.
Abstract: Objectives: To determine whether nurses with different clinical backgrounds make different triage decisions in NHS Direct, the 24-hour telephone helpline staffed by nurses.Methods: Mixed methods in...
TL;DR: As the most cost-effective use of resources, this work would advocate the use of double- rather than single-sided questionnaires, and use of a single-rather than multiple-booklet design.
Abstract: Objectives: Low response rates to postal questionnaires can threaten the validity of studies by reducing the effective sample size and introducing bias. The identification of methods with which to optimise response rates could, therefore, improve the quality of studies. In an attempt to identify such methods, we undertook a randomised trial of two simple variations in questionnaire design.
Methods: Using a 2 x 2 factorial design, we conducted a randomised trial to test two variations in questionnaire design; the questionnaires were printed on either single-sided or double-sided paper and had either a single- or multiple-booklet layout. Using equal random allocation, 3836 women were randomised to receive one of these questionnaires as part of a study investigating risk factors for osteoporotic fractures.
Results: One thousand eight hundred and seventy questionnaires were returned, giving an overall response rate of 48.7%. There were no significant differences in the overall response to each of the four questionnaire designs. When the number of responders who completed at least 50% of each of the three sections was identified, it was found that single-booklet questionnaires had a better response than the multiple-booklet questionnaires and that single-sided questionnaires had a better response than double-sided questionnaires. However, these results were not significant at the 5% level. There were no significant differences in the response to questions on the odd (left-hand side) pages for the single- compared with the double-sided questionnaires.
Conclusion: As the most cost-effective use of resources, we would advocate the use of double- rather than single-sided questionnaires, and use of a single- rather than multiple-booklet design.
TL;DR: This paper highlights key developments in the UK labour market, presents new findings from two nationally representative sample surveys, and advances some conclusions about the contradictory effects of change and continuity in paid work in the early years of the twenty-first century.
Abstract: Focusing on paid employment rather than work in its multiple forms, this paper highlights key developments in the UK labour market, presents new findings from two nationally representative sample surveys, and advances some conclusions about the contradictory effects of change and continuity in paid work in the early years of the twenty-first century. Reference is made to international developments to place the UK in a broader context.
TL;DR: The paper highlights the limitations in comparing national data on the nursing workforce, illustrating the extent to which currently collated national data can present a misleading picture of staff:population ratios in different countries.
Abstract: This paper focuses on one global aspect of the current health sector workforce policy agenda - the international recruitment and migration of health workers. It does so primarily by using a case study of the recruitment of nurses to the UK, as a means of exploring the policy challenges and associated research questions. The paper highlights the limitations in comparing national data on the nursing workforce, illustrating the extent to which currently collated national data can present a misleading picture of staff:population ratios in different countries. It then reports on the significant growth in the numbers of nurses entering the UK from other countries, using registration data. In 2001/02, more than 16 000 nurses entered the UK nursing register from non-UK sources. In this year, for the first time, the number exceeded the number of home-trained nurses. An analysis of postcode data highlights that these non-UK nurses have a younger age profile than home-based registered nurses and are more likely to report a postcode in London and south-east England. The paper also examines the push and pull factors that contribute to the international mobility of health workers. The paper concludes by examining the policy implications of this growing reliance on international recruitment, including the effect of the ethical guidelines on international recruitment introduced by the Department of Health in England.
TL;DR: Seven of the themes became the basis for a new clinical priority assessment criteria tool using visual analogue scales to determine priority of patients for elective general surgery.
Abstract: Objective: A single visual analogue scale is used in New Zealand to prioritise patients for elective general surgery. Although it reflects clinical judgement, it has been criticised for its lack of transparency. We wished to elicit generic criteria used by surgeons for prioritisation of patients for elective general surgery in order to improve the transparency of the visual analogue scale.Methods: Semi-structured interviews were undertaken with 15 general surgeons. Using the repertory grid method, surgeons were asked to explain their rationale for distinguishing between patients they considered a high, medium or low priority for treatment. Interviews were audiotaped, transcribed and analysed for themes. The accuracy of the thematic analysis was checked using a five-point Likert scale to assess surgeons' agreement with the identified themes. Further testing to check for face, content and construct validity was undertaken with a purposive sample of six surgeons prioritising patient vignettes.Results: Eight ...
TL;DR: Examining the social capital: health relationship revealed that with the exception of feelings of trust and reciprocity, no other social capital component made significant contributions to explaining health variance and that macro-level factors such as housing conditions and employment opportunities emerged as key explanatory factors.
Abstract: ObjectivesResearch into the social determinants of health inequalities is increasingly focusing on macro-level forces affecting individuals and communities. The concept of social capital has been a...
TL;DR: While the principle that non-physician clinicians can deliver high quality care within the practice of medicine is unequivocally true, more research is needed to test this principle under conditions of greater clinical complexity and autonomy, and caution must be exercised in applying this principle too broadly.
Abstract: Non-physician clinicians have become prominent providers of patient services within the practice of medicine. They include nurse practitioners and clinical nurse specialists, physician assistants, the alternative and complementary disciplines (chiropractic, naturopathy and acupuncture), mental health providers (psychologists, clinical social workers, counsellors and therapists) and specialty disciplines (optometrists, podiatrists, nurse anaesthetists and nurse-midwives). Although these various disciplines have differing histories and philosophic frameworks, which create distinctive approaches to patient care, they have shared a struggle to obtain recognition and autonomy through state licensure, to expand their state-granted practice prerogatives and to achieve broader reimbursement from third-party payers and managed care. Most entered into a growth spurt beginning in the early 1990s. All now provide care that not only overlaps that of physicians but that complements and supplements that care. The central question is, how does their care contribute to quality? The evidence thus far shows that non-physician clinicians throughout the range of disciplines can produce high-quality outcomes under particular circumstances. However, the strongest body of evidence is derived from care that is at the least complex end of the clinical spectrum or that is provided under the umbrella of physicians. Unfortunately, few studies have critically examined the outcomes of non-physician clinicians at the leading edge of their practice prerogatives and under conditions that are free of physician oversight. Thus, while the principle that they can deliver high quality care within the practice of medicine is unequivocally true, more research is needed to test this principle under conditions of greater clinical complexity and autonomy, and, pending the results of such research, caution must be exercised in applying this principle too broadly.
TL;DR: Variability in the use of CPAC tools meant that at the time of the study they did not provide a transparent and equitable method of determining access to surgery, and it is suggested that further development is difficult in the absence of evidence to identify patients who will benefit the most from surgery.
Abstract: Objectives: To describe the ways patients access elective surgery in New Zealand, and to understand the use of, and attitudes to, clinical priority assessment criteria (CPAC) in determining access ...
TL;DR: The development of the Specialist Outreach Service in the Northern Territory and its initial process evaluation is described, and a Cochrane systematic review is summarized to summarise the published research on the effectiveness of specialist outreach in improving access to tertiary and hospital-based care.
Abstract: People in remote Aboriginal communities in the Northern Territory have greater morbidity and mortality than other Australians, but face considerable barriers when accessing hospital-based specialist services. The Specialist Outreach Service, which began in 1997, was a novel policy initiative to improve access by providing a regular multidisciplinary visiting specialist services to remote communities. It led to two interesting juxtapositions: that of 'state of the art' specialist services alongside under-resourced primary care in remote and relatively traditional Aboriginal communities; and that of attempts to develop an evidence base for the effectiveness of outreach, while meeting the short-term evaluative requirements of policy-makers. In this essay, first we describe the development of the service in the Northern Territory and its initial process evaluation. Through a Cochrane systematic review we then summarise the published research on the effectiveness of specialist outreach in improving access to tertiary and hospital-based care. Finally we describe the findings of an observational population-based study of the use of specialist services and the impact of outreach to three remote communities over 11 years. Specialist outreach improves access to specialist care and may lessen the demand for both outpatient and inpatient hospital care. Specialist outreach is, however, dependent on well-functioning primary care. According to the way in which outreach is conducted and the service is organised, it can either support primary care or it can hinder primary care and, as a result, reduce its own effectiveness.