CARI - Caring for Australasians with Renal Impairment

Michelle Irving and co-authors undertook a systematic review on the topic of implementation of evidence-based medicine in chronic kidney disease (CKD). A comprehensive search of MEDLINE and EMBASE from January 1966 to June 2007 was undertaken. Twenty two studies, including 7 randomized controlled trials and 15 before-after studies were included. Four main interventions were evaluated in over 700 dialysis centres/hospitals or general practices: audit and feedback, computerized decision support system (CDSS), opinion leader/multidisciplinary team and passive dissemination of guidelines. Due to heterogeneity, a formal meta-analysis was not possible.

Audit and feedback significantly increased 14 of the 25 study outcomes with a median improvement of 2.5% (range: -4.5-48.4%). CDSS significantly increased 3 of the 4 study outcomes with a median improvement of 12.8% (range: 1.1-42.1%). Opinion leader/multidisciplinary team significantly increased 24 of the 30 study outcomes with a median improvement of 8.2 %( range: -4.0-79.8%). Dissemination of guidelines resulted in a median improvement in study outcomes of 2.7% (range 0.5-25.8).

We found that well planned and executed interventions were able to improve CKD management to varying degrees. More active and multifaceted interventions were more effective than passive dissemination and single strategy methods. Interventions need to address all barriers to implementation and provide adequate workforce assistance. Specific quality indicator targets were associated with greater improvements in care.

Survey of nephrologists

In 2006, a self-administered survey was distributed to all Australian and New Zealand nephrologists. Seven questions were repeated from a similar survey held in 2002. A total of 211 nephrologists (70% of practising nephrologists) responded. Over 90% agreed that the CARI Guidelines were a useful summary of evidence and nearly 60% reported that the guidelines had significantly influenced their practice. The proportion of nephrologists reporting that the guidelines had improved patient outcomes rose from 14% in 2002 to 38% in 2006. The proportion of nephrologists indicating that the guidelines did not match the best available evidence decreased from 30% in 2002 to 8% in 2006. Almost all nephrologists regarded the CARI Guidelines as useful evidence summaries, although only one-third believed that guidelines affected health outcomes. Attitudes to the guidelines have become more favourable over time; this may reflect changes in the CARI process or attitudinal changes to evidence among nephrologists.

More details regarding the results of this study will be published by the AJKD in 2009.

Survey of renal nurses

A self-administered survey was distributed to all members of the Renal Society of Australasia in 2006. The results were compared with those from a similar survey run in 2002. Of the 173 respondents, over 95% considered the CARI Guidelines to be a good synthesis of the available evidence; 80% indicated that the guidelines had significantly influenced their practice; and 86% considered that the guidelines had improved outcomes for patients. Although nephrologists were generally positive regarding the CARI Guidelines, renal nurses were more positive, especially regarding the effect of the guidelines on practice and the improvement in health outcomes. Australian and New Zealand renal nurses valued the CARI Guidelines highly, used them in practice, and considered that they led to improved patient outcomes. Positive responses towards the guidelines increased between 2002 and 2006.

A paper outlining further results of this study have been submitted for publication.

Qualitative study: Understanding the role of guidelines in clinical decision-making and clinical practice

Semi-structured, face to face interviews were undertaken with 19 nephrologists purposively sampled from a variety of clinical settings across Australia. Participants were asked about their views on nephrology clinical practice guidelines, which included the CARI Guidelines, and international guidelines and other influences on their clinical practice. Four major themes were identified: 1) trust in the CARI process and output; 2) the function of guidelines in clinical practice; 3) non-guideline influences on clinical decision making; and 4) facilitators of guideline implementation. Conclusion: Evidence-based guidelines strongly impact on the clinical decision-making of Australian nephrologists but are only one input. Improvements in the evidence which underpins guidelines and improvements in the content and formatting of guidelines are likely to make them more influential on decision making.

Qualitative research – consumer preferences

The inclusion of consumer preferences in prioritizing research topics is widely advocated, but prioritization is driven largely by professional agendas. Patients with chronic kidney disease (CKD) were purposively sampled from four kidney dialysis and transplant centres in Australia to participate in nine focus groups (three each for pre-dialysis, dialysis and transplant patients), which were conducted from July 2006 to September 2006. Each involved 6-8 participants. Transcripts were coded and thematically analysed to identify recurrent research topics and the participants' reasons for their choices. Participants suggested 8 research priorities: prevention of kidney disease, better access to and improvement in kidney transplantation, reduction of symptoms of CKD and complications associated with treatment, new technological therapies, psychosocial aspects of living with CKD, whole body not organ-specialized care, and improvement in dialysis and caregiver support.

Five major reasons for the selections were identified: normalization of life (developing therapies and regimens that fit into daily living), altruism (considering the welfare of others before personal needs), economic efficiency (channelling resources for maximum economic gain), personal needs (preferences based on feelings, values, personal needs) and clinical outcomes (improving health states and the physiological condition of patients with CKD). A patient-focused research agenda is possible to elicit for CKD, and by inference for other healthcare issues. Unlike researchers who focus on specific interventions and questions, consumers think in terms of broad themes and quality of life outcomes. Effective methods for translating a patient-focused agenda into research priority setting and resource allocation are now needed.

Comparison of development methodologies used by guideline groups

There are calls for guideline organisations to pool their resources and develop guidelines collaboratively rather than individually. To achieve this, the methodological principles used by groups would need to be common and result in the systematic identification and synthesis of the best available scientific evidence. Denise Campbell and co-authors compared the development process of 3 nephrology guideline groups with the process used by a number of major guideline organisations. The latest NHMRC guidance on essential steps in guideline development was used as a methodology benchmark. A sample guideline from each organisation was also assessed for quality, using the AGREE criteria.

The results showed that the quality of guidelines produced by non-nephrology organisations is generally higher than those produced by nephrology groups. There are wide variations in the process used by different groups and some of these are due to the difference in resources and skills available to a group. Few organisations assess the impact of socioeconomic position in relation to the condition of interest, determine benefits/harms, compare costs/benefits of interventions, or determine cost-effectiveness and feasibility of recommendations as a routine part of guideline development. Most organisation’s guidelines yield low ’Applicability’ scores, which means Work Groups are not considering the likely organisational, behavioural and cost implications of applying a guideline. There is a need for guideline organisations to expand their methodological process so that more thoughtful and practical guideline recommendations are developed. A poster of the research was presented at the Guidelines International Network conference in Toronto, Canada in August 2007.

Trial-based guidelines and chronic kidney disease

Ideally, clinical practice guidelines are supported by high quality evidence but trials in chronic kidney disease are relatively few in number compared with other specialties in internal medicine. Allison Tong and co-authors assessed the proportion of CARI Guideline Recommendations supported by high quality evidence compared with Suggestions for Clinical Care, based on low or missing evidence. All CARI Guidelines published from 2004-2006 were reviewed.

Analysis of the data showed that 43.4% of guideline subtopics were supported by at least Level I or II evidence (i.e. systematic review, randomised controlled trial), 34.1% were supported by Level III or IV evidence (i.e. cohort studies, case control studies etc.) and 22.5% were not supported by any trial evidence. This applied across the 3 topic areas – chronic kidney disease, dialysis and transplantation. The conclusion was that there is a need for more high quality RCT evidence in nephrology, which will in turn lead to the development of more valid clinical practice guidelines. A poster of the research was presented at the ANZSN Annual Scientific Meeting in September 2007 at the Gold Coast and at the Guidelines International Network conference in Toronto, Canada in August 2007.