Summary

In a recent symposium “Primary care for Hong Kong–the way forward”, the usefulness of clinical practice guidelines in improving health outcomes was debated. In this paper, we discuss the evidence on whether implementing clinical practice guidelines will improve health outcomes in primary care, the attitude of primary care doctors towards guidelines and the factors that may facilitate primary care doctors’ adherence to evidence-based guideline recommendations.

摘要

近期，在「香港基層醫療服務前瞻」研討會上，藉著實行臨床醫療指引以改善醫療後果這個方法的效益備受爭論。本文討論使用臨床醫療指引來改善基層醫療後果的實證，基層醫生對待指引的態度，和有助基層醫生遵守實證指引建議的因素。

---

Introduction

In the recent symposium “Primary care for Hong Kong - the way forward”, Professor Barbara Starfield presented ecological data to show that in many countries, primary care is important in achieving better health outcomes with lower costs and with greater equity in health.¹ She introduced the Patient-Centred Medical Home (PCMH) as one of US’s care models to enhance primary care, emphasizing electronic health records, team work and chronic care guidelines. However, she raised one question, “To what extent is the PCMH’s disease-oriented approach consistent with the principles of population-based primary care?” She argued that “what is needed is person-focused care over time, not disease-focused care”.²

As clinical practice guidelines (CPGs) mostly focus on single diseases, their usefulness in improving quality of primary care was debated in the symposium. On the one hand, “what characterizes illness is its variability, not its average manifestations... [but] randomized controlled trials are based on the average response”. On the other hand, “we have varying patterns of care, so should we be looking at guidelines...? I would argue that guidelines are useful at a system level. We need to understand common standards and common practices.”³

The lively discussion in the symposium led to a few follow-up questions:

1. What is the evidence that CPGs improve health outcomes in primary care?
2. What is the attitude of primary care doctors towards CPGs?
3. What may facilitate primary care doctors’ adherence to evidence-based guideline recommendations?

In this paper, we summarize the literature search for the above questions, and discuss the areas that may need further study for CPG development in Hong Kong’s primary care.

Do clinical practice guidelines improve health outcomes in primary care?

Grimshaw et al’s paper is the first systematic review on CPG effectiveness: out of 59 evaluations, most showed improvement in the process or outcome of care, although the size of the improvements varied considerably.⁴ However, Worrall et al’s systematic review, which focused on the impact of CPG on quality of primary care concluded otherwise: only 5 of the 13 trials produced statistically significant results.⁵ Some before and after studies did show improvements (for example, decreasing emergency department visits in asthma patients), but were limited by the study design.⁶ In a recent systematic review on the impact of CPGs in the Netherlands, 17 of the 19 studies reported improvement in the process and structure of care, while six out of nine studies showed small but significant improvements in health outcomes (for example, HbA1c lowered by 1.7%, body weight decreased by 3.8kg, incidence of pressure ulcers reduced from 54 to 32 per 1000 patient days).⁷ Hence, there is some evidence that planned and thoughtful implementation of high quality CPGs does improve the structure, process and outcome of primary care in a defined population.

In his 2009/2010 policy address, the Hong Kong Chief Executive pledged to provide additional resources for developing chronic disease management protocols. What could Hong Kong learn from other countries? It has been observed that the volume of CPGs grossly outnumbers studies that assess their effectiveness.⁷ At the very least, evaluation on effectiveness in improving quality should be an integral component in future CPG implementation.

What is the attitude of primary care doctors towards CPGs?

Farquahar et al summarized 30 high response (>60%) surveys in the Netherlands, US, UK, Italy, Canada, Australia, Denmark, Ireland and Israel, assessing clinicians’ attitudes to CPGs.⁸ Among the 11,611 respondents, about two-thirds were primary care doctors. They found that most doctors in the surveys were supportive of CPGs, finding them to be useful, educational and likely to improve quality of care. It is noteworthy that nearly half of respondents showed concern over the increased likelihood of litigation or disciplinary action.

Qualitative methods often allow deeper exploration of attitudes. Carlsen et al reported 12 focus groups or semi-structured interviews with general practitioners/family physicians (GP/FPs) in the UK, US, Canada and the Netherlands.⁹ The themes identified highlight the value the primary care doctors place on seeing every patient as a unique individual. Many of the doctors argued that population-based trials were not applicable equally to all, and were skeptical about the `one-size-fits-all’ prescription in most CPGs.

No studies have been done to explore the attitude of Hong Kong’s primary care doctors towards CPGs. Information on this area would be important and relevant for local guideline development policy.

What may facilitate primary care doctors’ adherence to guideline recommendations?

The EPOC Cochrane Review Group has reviewed a series of methods of implementing CPGs, but none is reliably effective in all settings and circumstances.¹⁰ A study on Dutch GPs revealed that within each guideline, each recommendation has a unique pattern of barriers,¹¹ hence the authors suggested that careful analysis of barriers among end-users would be paramount, and implementation strategies should be tailored and barrier-driven.

CPGs can be developed using centralized and decentralized approaches. At the moment, Hong Kong does not have any central agencies for CPG development or implementation. The Hong Kong College of Family Physicians drafted a few management guidelines in 1991^12-14 but the effort was not sustainable after two years. Under the Professional Development and Quality Assurance, Department of Health, there are currently CPGs on pertinent primary care conditions like type 2 diabetes mellitus, hypertension and lipid management.¹²  A couple of CPGs on conditions like osteoarthritis,¹³ epilepsy¹⁴ or bipolar disorder¹⁵ have been developed and disseminated in a haphazard manner by different regional organizations. Clinician’s guides on some common problems are recently accessible from the Family Medicine Unit website of The University of Hong Kong.¹⁶

In a recent JAMA editorial,¹⁷ misuse of evidence in the revised American College of Cardiology/American Heart Association guidelines was discussed.¹⁸ The editors called for creating codes to govern the development process and limit guideline panel membership only to those with no conflict of interest. They went on to suggest that “the time has come for guideline development to again be centralized... such centralization should help reduce bias and redundancy, and better guide the research agenda.” It is not known whether Hong Kong’s primary care doctors support the centralized approach for CPG development. Nevertheless, several studies have shown that involving primary care doctors in the process of developing CPGs will enhance their use in daily practice. ^19-20

Lastly, the use of economic incentives for guideline implementation is controversial. It has been shown to be favoured by British GPs²¹ but other countries (for example, Norway) did not consider financial reward a suitable method for promoting guideline implementation.¹⁹ Whether Hong Kong should follow the British model is open for discussion.

It is interesting to note that these initiatives to improve guideline adherence all involve interventions focused on primary care practitioners. Perhaps we also need studies that evaluate the applicability of a newly produced guideline in routine primary care. Such studies would increase our understanding of the challenges primary care doctors might face in applying recommendations to the care of the individual, and of balancing the needs of managing several clinical conditions present in the same patient.²² If we can understand the issues from the primary care doctor's perspective, perhaps guideline development groups could be assisted in framing recommendations that more often account for the perceived problems of routine practice.

Conclusion

If implemented, evidence-based CPGs can potentially improve quality of care, and evaluation on effectiveness should be integral in their promotion. Primary care doctors worldwide are supportive of CPGs but further study is needed to explore local attitudes and barriers, as well as strategies that are most likely to have an impact on facilitating adherence to evidence-based guideline recommendations in Hong Kong's primary care setting.

Key messages

1. There is some evidence that planned and thoughtful implementation of high quality clinical practice guidelines improves the structure, process and outcome of primary care in a defined population.
2. Doctors worldwide are supportive of clinical practice guidelines, finding them to be useful, educational and likely to improve quality of care; however, the attitudes of Hong Kong’s primary care doctors towards guidelines are not known.
3. Financial incentives, barrier-driven implementation strategies and end-user involvement in the process of development may improve adherence to guideline recommendations; yet, it is not known whether these factors will have an impact on guideline adherence in the local primary care setting.
4. In particular, a centralized approach for guideline development may help reduce bias and redundancy, but whether Hong Kong’s primary care doctors support the centralized approach is to be explored.
5. More studies are also needed to understand challenges primary care doctors may face in applying recommendations to the care of the individual, and of managing multiple clinical conditions in the same patient, so that recommendations could be framed to accommodate perceived problems of routine practice.

---

---

References