Summary

General Practitioners (GPs) encounter questions in the day-to-day clinical work but many of these questions are left unanswered. Evidence-based medicine (EBM) is a paradigm that can help GPs to meet their information needs, make better clinical decisions and improve patient outcomes. GPs worldwide have mixed feelings towards the increasing profile of EBM. Main obstacles cited include a negative attitude, lack of time, access problems and deficiency in skills in searching and critical appraisal. New developments in the model of EBM, improvements in medical informatics and practical suggestions that may help promote the learning and practice of EBM are discussed from the perspective of the local context.

摘要

家庭醫生在日常臨床工作遇到的問題，有很多都是難於解答的。循證醫學作為一種範例可讓醫生獲得所需要的醫學資訊，從而作出更佳的臨床決定，提高醫療水準。世界各地的家庭醫生，對循證醫學的冒起都有不同的看法和感受。所引述的主要障礙包括：對循證醫學的理念持否定態度，時間不足，存取上的困難及不懂得如何檢索或評估證據。本文針對以上的問題，總結了循證醫學在概念及資訊科技上的新發展，盼能藉此促進本港家庭醫生對循證醫學的學習及實踐。

---

Introduction

This patient's blood pressure has been well-controlled for a while, can I see her less frequently, say, every 6 months? To save time, can I take blood pressures without rolling up patients' sleeves? When should I take a follow-up XR for this elderly man who has been treated for pneumonia? For Chinese patients, is a lower dose of lipid-lowering drug just as effective? How sure am I when I tell my patients that treatment of Helicobacter pylori infection can reduce the risk of gastric cancer? Can I save money by using plain soap if it is as effective as anti-bacterial soap in removing pathogenic agents from my hands?

EBM is the "conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients".¹ Asking questions is often considered to be a constructive way of making clinical decisions. If some questions arise within our daily clinical context, we can proceed with the following, as suggested in Table 1.

Do GPs encounter questions during patient consultations?

The answer is yes. Various studies showed that General Practitioners (GPs) asked five questions per patient to 0.5 questions per half-day, depending on the setting and research methodology. More questions were raised in a teaching setting or under the influence of prompting during a study.² The three most common types of questions were "What is the cause of symptom X?" "What is the dose of drug X?" and "How should I manage disease or finding X?"³

From various studies, most of these questions remained unanswered.^3,10 The likelihood of pursuing a question is increased if the problem was perceived as urgent or when a definitive answer was thought to exist.³ Belief that the patient expected the answer or fear of malpractice exposure were also motivating factors.¹¹ In one study, the sources of answers were textbooks (31%), journal articles (21%) or senior colleagues (17%).⁴ GPs seemed to make more use of printed sources of information than electronic databases.⁵ The information is likely to be neither the best nor the most up-to-date.

How do GPs feel about EBM?

It takes effort and practice to acquire the skill in articulating a focused clinical question. Framing well-built questions consisting of patient information, intervention, comparison and outcome (PICO) is an effective way of translating information needs into answerable questions.⁶ However, this step-by-step way of "asking, searching, appraising and applying" the best current evidence engenders both negative and positive reactions from clinicians.⁷ Questionnaire surveys and qualitative studies from Australia,^8,9 UK,^10,11 Canada,^12-14 USA,¹⁵ New Zealand¹⁶ and Saudi Arabia¹⁷ showed that GPs have mixed views about the use of this method in practice.

To summarize, the most commonly cited obstacles include the following:

1. Scepticism towards the philosophy, usefulness and applicability of EBM;
2. Lack of time;
3. Access problem to evidence at the point of care;
4. Knowledge and skills in searching and appraising evidence.

EBM and the art of medicine in primary care practice

GPs in general, believe that EBM improves patient care and is useful in helping clinical decision making and patient management.^8,9,12-14,17 It could also be a helpful strategy for meeting their information needs.^16,17 However, some GPs find the EBM methodology a devaluation of "the art of medicine",¹² especially in primary care practice where intuition or clinical hunch plays a vital role.¹⁰ Some feel that patients' demand or expectations and the practice setting exert a more powerful influence on clinical decision making than pure research findings.^12,13 The complexity in implementing evidence in "real-life individual patients" often give rise to the experience of conflicts within the doctor,^8,12 or it may jeopardize the doctor-patient-relationship.¹⁰ Trustworthiness and credibility of evidence is of great concern, especially with respect to the influence of the pharmaceutical industry.^12,15 Lack of relevant research evidence in primary care is frustrating to those who make fruitless attempts to pursue an answer.¹⁵ Some quotations from GPs citing their views towards EBM are listed in Table 2.

An updated model in EBM

Since the promotion of EBM in 1992,¹⁸ there has been hot debate on the legitimacy of the above criticisms. In fact, significant modifications to the original model have been made by the founding fathers, indicating that EBM itself is an evolving methodology currently in transition. In the newly-articulated framework, the interaction among clinical expertise, patient preferences/values, research findings are highlighted. To quote, "EBM was developed to encourage practitioners and patients to pay due respect - no more, no less - to current best evidence in making decisions".¹⁹ (Figure 1) As Greenhalgh²⁰ pointed out, the "zero-sum" relationship between the deductive steps of EBM and the subjective interpretation of the patient's story in clinical encounters is a myth and a false dichotomy which should be acknowledged and overcome. Fitzpatrick²¹ observed that the rigid insistence on controlled trials as the sole source of evidence on effectiveness, as advocated in the beginning of the EBM movement, is fading. Researchers increasingly use particular study architectures (such as "N-of-1" and large, simple trials) and carry out judicious subgroup analyses to improve the extrapolation of research results to individual patients in the "real world". Qualitative research is becoming recognized as an important source of evidence as contextual factors are given more weight compared with quantitative research. The paucity of consistent and high quality evidence in the primary care setting is globally recognized.²² The exponential growth in clinical research, coupled with international efforts to identify, sort and rationalize the evidence systematically, will hopefully close many of these gaps.

What is the situation in Hong Kong?

There is no local data on the attitude/beliefs of Hong Kong GPs towards EBM. Modules on EBM have been incorporated into medical undergraduate curriculum and diploma courses in Family Medicine. Yet, to date, less than 1.5% of active members in our college have fulfilled quality assurance criteria through engaging in EBM (Critical Appraisal) activity. (18 and 23 members in 2002 and 2003 respectively, unpublished data). The views expressed worldwide are in line with the author's experience during peer group learning and small group teaching. As Hazlett,²³ organizer of the first local EBM workshops in 1997, rightly pointed out, "If the utility of EBM is viewed with scepticism, it is highly unlikely that a clinician's decision making will become increasingly dependent on the EBM rules of evidence and protocols". It is hoped that a more inclusive view of evidence as stated above will address some of the concerns raised by GPs with respect to the philosophy, usefulness and applicability of EBM.

A highly rated barrier in practicing EBM - Lack of time

How can busy GPs overcome the lack of time amongst a heavy patient load? In one study, family doctors spent less than 2 minutes seeking an answer to a question.³ Practicing doctors do not have the time to search multiple sites or scroll through long text. It has been shown that personal collections (textbooks, journals, and notes) remained the preferred information source.²⁴ Electronic searching ranked second.

The need of doctors to pick up real-time evidence which has been digested into quickly accessible summaries has been acknowledged and addressed. Electronic searching for the best current evidence is now cut to a few seconds, thanks to the explosive developments in medical informatics. New automated methods were developed to simultaneously search for multiple databases like the Cochrane Library/National Guideline Clearing house/PubMed. Examples of such meta-searching service include SUM search²⁵ and TRIP database²⁶ (Table 3). Even single-word entries of large topic areas such as hypertension or diabetes lead to one page summaries which "drill down" the medical literature in a hierarchy according to relevance and validity. User-friendly prompts are built in to guide the inexperienced users to refine their questions and search. Links to resources that provide broad discussion: relevant textbooks, traditional review articles and practice guidelines are also provided. This feature is most valuable when we are confronted with new/rare problems and feel crippled in articulating a focused question.

A list of useful EBM websites can be found in Table 4. Building personalized portals to home pages with links to these EBM websites is highly recommended. In Hong Kong, the Professional Development and Quality Assurance (PDQA) from the Department of Health is an accessible local site with user-friendly information for starters.²⁷ An important time and energy-saving solution for busy clinicians is to make full use of high quality pre-appraised evidence. Examples include evidence-based journals (Evidence Based Evidence, Best Evidence, Bandolier) which contain abstracts of quality- and relevance- filtered studies, data-bases of POEM²⁸ (patient-oriented evidence that matters) and CAT²⁹ (critically appraised topics - 1 page summaries of evidence relevant to common clinical questions). These high quality resources transform original research to succinct evidence from which clinicians can obtain information regarding outcomes that matter most to patients.

Access to high quality evidence at the point of care

"Clean drinking water" was the analogy Glasziou used when the Cochrane Library was available free to all Australian in 2002.³⁰ In fact, many of the databases mentioned above are not free of charge. It is hoped that free or less expensive access to "clean health information" like the Cochrane Library could be realized locally in the near future. For the time-being, honorary teachers for medical students have the privilege of gaining access to full-text articles via the Medical Library Quick Links from HKU.³¹ e-KG (e-Knowledge Gateway) is a website from the Hospital Authority providing full-text primary and secondary appraised e-journals and access over 20 medical database. Extension of this service at a reasonable subscription rate to doctors in the community is underway.³²

In a local study on physicians' attitudes towards the effects of computers on health care, the two highest ratings on the 17-item instrument were: providing clinicians access to current evidence and continuing education.³³ Benefits of computers and the worldwide web in keeping physicians up-to-date are well known. However, in a population-based survey among local doctors in 2000, only about half of the doctors had computerized their practices. It was described as "alarming" for solo or small group practices in which 70% had yet to computerize any clinical function. These groups of solo or small group GPs, who are responsible for 85% of ambulatory care, also lag behind their colleagues in the implementation of computerization at one year follow up.³⁴ Time costs, lack of technical support and capital investments were the biggest barriers to computerization.³⁵ A positive attitude and better computer knowledge were incentives to implement change, especially for those in individual practices, as they have an edge on the locus of control in making decisions on their own clinics.

How can GPs acquire the knowledge and skills to practice EBM?

Medical informatics education is important in promoting more positive attitudes and increasing computer knowledge amongst local GPs. Apart from computer skills, the mastery and application of critical appraisal skills are formidable tasks and should not be underestimated. In a survey on Australian GPs,⁹ 30% did not understand the term "systematic review", and only a minority understood the terms "relative risk" (23%), "absolute risk" (28%) and "number needed to treat" (15%) sufficiently to explain to others. Guyatt proposed that "not all clinicians need to appraise evidence from scratch but all need some skills".³⁴

Contrary to the "just-in-time" application of evidence at the point of care, protected time is more desirable to learn about the various skills involved in the practice of EBM. Attending workshops,²⁹ problem-based small groups,³⁵ e-mail discussion groups³⁶ will be time well-spent. It is beyond the scope of this short paper to discuss the practice of EBM in depth: excellent journal articles, books, websites are abundantly available for self-study or group use (Table 4). On top of searching for the best evidence at the point of care, an alternative strategy for busy GPs is to keep a log of questions in a sheet pad that looks like a prescription pad.³⁷ Evidence-based answers to many of the clinical questions can be pursued when surgery is over on one's own or, more preferably, with a group. Searching of the questions posed in the "Introduction" of this article has been done within a few minutes. Interesting "answers" to those questions can be found in the references.^40-45

Conclusion

Every encounter with a patient identifies gaps in our knowledge about the aetiology, diagnosis, prognosis, or therapy of their illness. Cultivating an enquiring attitude is a step towards the practice of EBM, which will improve health outcomes of patients. Many new developments have made the evidence readily available; it is just up to us to ask, to search, to appraise and to apply the best current evidence in daily patient care. The concepts and skills of EBM are promising to improve primary care in many different ways, one of which is helping front line practitioners to actualize the ideal of lifelong self-directed learning.

Acknowledgement

Acknowledgement is made to sincerely thank Dr Cindy Lam for her constructive comments and invaluable advice in this article.

Key messages

1. General Practitioners (GPs) encounter questions in the day-to-day clinical work but many of these questions are left unanswered.
2. Evidence-based Medicine is the "conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients". It is a paradigm that can help GPs to meet their information needs.
3. Main obstacles in the practice of EBM include a negative attitude, lack of time, access problems and deficiency in skills in searching and critical appraisal.
4. New developments in the concept of evidence within the EBM model, introduction of international and local educational resources/electronic databases and tactics of using small group/e-mail discussion/EBM prescription pads are suggested to overcome the barriers cited.

---

---

References