|HOME||HELP||FEEDBACK||BROWSE ARTICLES||BROWSE BY AUTHOR|
clinmed/2000070008v1 (August 21, 2000)
Contact author(s) for copyright information
Clinical guidelines: A means to too many ends?
1Rebecca Lawton (PhD)
School of Psychology
University of Leeds
Leeds, LS2 9JT
Tel: 0113 233 6641
Fax: 0113 233 5749
2 Anthony Burton (PhD)
Nuffield Institute for Health
University of Leeds
Leeds, LS2 9JT
Tel: 0113 233 3419
1 Contact for correspondence concerning the manuscript and requests for reprints
(Running head: Process controls in the NHS)
Clinical protocols, clinical guidelines, standards, rules – these terms are proliferating as fast as the written documents. These regulations are produced in the name of risk management, quality improvement, evidence based medicine, reducing practice variations and cost control. To further exacerbate the situation there is no direct mapping of terminology with objectives. The increase in these forms of process control in the health care sector shows no sign of abating. In this paper we use evidence from other industries to discuss whether these many objectives can be achieved through the implementation of clinical guidelines.
Given the complexity inherent in these objectives, it is understandable that the government has attempted to distil them into one policy aim: Implementation of clinical governance, defined as:
"A framework through which NHS organisations are accountable for continuously improving the quality of their services and safeguarding high standards of care by creating an environment in which clinical care will flourish." 
The last part of this definition leaves open the question of how the proposed framework should be implemented. However, implementing clinical governance has become synonymous with creating process controls, whose purpose is to regulate the actions of clinicians (see for example , , , ). We contrast this with outcome controls where regulation is achieved by assessing the after-effects of actions. 
In this paper we address the question of whether, and under what conditions, process controls are the best form of organisational control. Applying the experience of other industries to the health service we will argue that a range of regulatory strategies are needed to achieve the goals of clinical governance. Further we will suggest that the most effective organisations are those that facilitate learning within their systems. 
Why process controls?
In the white paper on clinical governance there is a curious switch in emphasis. In the introduction, where the aims of the government policy are discussed, the emphasis is on improving outcomes. But in the bulk of the white paper, on the practical measures for achieving this improvement, the focus becomes processes to be implemented. It is striking that so little attention is paid to outcomes, other than monitoring the successful implementation of procedures. For instance, in the chapter ‘Monitoring quality Standards’, the question ‘How will the Commission for Health Improvement Help patients?’ is answered with statements on providing ‘reassurance that effective systems are in place’, ‘working to reduce variations in services’, ‘providing feedback into the National Service Frameworks’ and ‘monitoring of uptake of NICE guidance’. 
Thus, monitoring the success of process controls in improving the quality of outcome becomes a key issue. Currently the balance of evidence is in favour of guidelines but remains inconclusive (, ). For example in a review of 91 rigorous studies of the implementation of clinical practice guidelines, Grimshaw et al. conclude that "properly developed guidelines can change clinical practice and may lead to changes in patient outcome"(authors’ italics) . Moreover, taken individually, five of the seven studies showed no effect of adherence to guidelines on patient outcome. In fact the majority of studies (70 out of 91) investigating guideline implementation were concerned only with the effect this had on the process of care e.g. does practice change or do doctors comply. What is clear from this and other research is that the dissemination of guidelines alone cannot be assumed to alter practice, or indeed to improve patient outcomes.
Even more fundamental to a discussion of process controls is the extent to which it is feasible to proceduralise a task. Guidelines are costly and time consuming to prepare, and so, it has been argued, are only appropriate for common conditions or when they have a substantial impact on cost or outcome. Guidelines that cannot be audited are of limited value. 
Where the costs of producing the guidelines are low, Perrow argues there are two important aspects that influence the extent to which a task may be usefully pre-programmed.  The first is the number of ‘exceptional cases’ – new events or problems encountered that are perceived as unfamiliar. The second is the nature of the ‘search process’ undertaken when these exceptions occur. Are the problems solvable via logical analytical steps or do they require experience or intuition for a solution? If there are many, largely unanalysable, exceptions then it becomes difficult to develop procedures to govern performance. At the opposite extreme, routine jobs, with few exceptions and analysable problems are easy to proceduralise.
Further, it may not always be desirable to implement feasible procedures. The advantages and disadvantages of prescriptive rules are well known and will be true of clinical guidelines. For example Hale and Swuste report that safety rules avoid the need to continually ‘reinvent the wheel’, promote predictability and uniformity among members’ behaviour and, so it follows, define responsibilities. However, excessive prescription can cause blindness to new situations, resentment at the loss of freedom, repress innovation and learning in the system and place high demands on management for generation and supervision. 
The experience of other industries leads us to predict that guidelines will be a useful tool where there is little patient variation (few exceptional cases) and a strong evidence base (the problem is analysable). Where there is much patient variation and the problem is analysable only by the doctor on the spot, guidelines are less appropriate. A useful analogy here is aviation. For the routine take-off, flight and landing, the pilot may be functioning largely at a skill and rule based level i.e. following checklists and procedures. However, in a novel situation the experienced pilot will use a greater number of situational cues in reaching a decision, using previous experience and training to determine the relevance of a particular cue. "This enables them to employ checklists and procedures as guidelines rather than as absolute prescriptions of courses of action" (p.85). 
Guiding or directing?
If for the moment we accept that process controls are the best means of enhancing outcomes, and are employed only where feasible, there is still ambiguity in the terms used to refer to the control of process. These include protocols, standards, care pathways, rules and policies. It is unclear to healthcare professionals how a clinical guideline differs from and relates to these other forms of control (, ). For the practitioner the question becomes how prescriptive is the guidance meant to be?
Hale and Swuste propose that rules can be of three types that limit the freedom of the actor to varying degrees.  The first type of rule defines the goals to be achieved (e.g. zero accidents); the second type defines the way in which decisions about a course of action should be taken (e.g. refer to manual 10) and the third type defines a concrete action or required state of the system (e.g. safety goggles must be worn). All three types of rule are present in the National Service Framework’s ‘Protocol for the prevention of coronary heart disease in practice’. The protocol’s first aim is ‘to identify all patients with established cardiovascular disease and offer them comprehensive advice and appropriate treatment to reduce their risks’ (type one rule). The full protocol is an example of the second type of rule since it offers a means of deciding which patients are at-risk. Finally the protocol includes specific actions required in the diagnosis of patients e.g. blood pressure should be measured in the sitting position after the patient has been rested for 5 minutes. This is an example of represents the third type of rule, defining the concrete action required.
In using this protocol as an exemplar we are criticising neither the protocol itself nor the National Service Framework. Instead we hope to draw attention to the fact that each type of rule requires different strategies for implementation and monitoring, and different behaviour on the part of the doctor. A type 3 rule limits freedom and requires compliance, which is checked by monitoring behaviour. At the other extreme a type 1 rule leaves the detailed course of action fairly open and control is achieved through monitoring outcomes.
The use of the term guideline may be a political attempt to satisfy doctors’ desire for autonomy, yet some components of this protocol, are not guidance, but direction. This is common to many industries where what is required is compliance with prescriptive rules. In many of these industries there are well established systems for monitoring and enforcing type 3 rules (e.g. training and testing of rule knowledge, supervisory presence, disciplinary procedures). In others, e.g. aviation, there is little need for external regulation as pilots have largely internalised the need to comply with type 3 rules and therefore self- regulation works.
There are two questions that healthcare professionals need to address. First, to what extent should guidelines include type 3 rules. For the NHS, it is important to acknowledge that rules are frequently of type 1 and 2 and, as such, the optimal level of compliance may not be 100%. For instance if there are contra-indications it may be inappropriate to comply with a guideline. Compliance with a rule is not always good and violation is not always bad, and the decision depends very much on the appropriateness of a rule for a given situation .
Second, to what extent should external monitoring play a role in the implementation of rules. For type 3 rules, if internalisation of the values of the organisation occurs, there is a lesser requirement for external monitoring of compliance. At present there is little evidence to suggest that internalisation of organisational values is occurring. Consultants are reportedly those who are most opposed to clinical guidelines .
For rules of type 1 and 2, guidance rather than direction, the emphasis is on monitoring of outcomes rather than process. In answering the present public concern with the regulation of healthcare professionals, we need to deal with these issues.
In reality, the implementation of clinical guidelines is proving a difficult task. For example, Grilli and Lomas reviewed 23 studies and report a mean compliance rate of 54.5% with clinical guidelines.  Again we can draw comparisons with other industries. Safety rules, the nearest equivalent of guidelines in an industrial setting, are frequently disregarded (, , ,.There is evidence that deviations arise where there is: 1) lack of knowledge or understanding of the rules; 2) another priority e.g. finishing the job; 3) no support for compliance at a local level e.g. through providing the necessary skills training, equipment, staff. , .
These issues apply equally in the health service. International studies examining factors such as familiarity with, confidence in and attitudes towards practice guidelines confirm the contentious nature of guidelines. (, , , ) Although considered useful in establishing best practice and as an educational tool, they represent a threat to professional autonomy. Moreover, knowledge of, and a positive attitude toward, guidelines does not guarantee compliance.  In the NHS, as in other industries, non-compliance may reflect a lack of support for clinicians. In a recent paper Hewison and Wilson (1999) describe the case of an obstetric guideline advising a ventouse rather than a forceps delivery. Obstetricians appeared reluctant to change their practice and compliance was low – why? Doctors were not given any training in the use of a ventouse, and those who had used forceps successfully predicted that they might have worse outcomes using the new technique.  Even with the support provided through skills training, it is possible that in the short run, patient outcomes might be worse. Thus, support for compliance in this case means much more than providing the initial skills training.
As Westrum argues, effective organisations have members who are empowered and have a ‘licence to think’.  Within such organisations, errors and incidents are widely reported, analysed and acted upon. Responsibility is shared, as the system freely admits and profits from its mistakes. This has not been true of the NHS and as a result, self-regulation of the medical profession is becoming increasingly unpopular with the general public. More effective monitoring of the outcome of care through audit, case review, incident reporting and near miss reporting systems, carried out stringently could achieve the control desired in many areas of medicine without the limitations on behaviour that accompany process control. There is a fine balance between restricting practice unnecessarily and reducing variation, between blind obedience and considered compliance.
1 Secretary of State for Health, A First Class Service: quality in the new NHS. London: Dept of Health, 1998
2 Chassin M. Practice guidelines: best hope for quality improvement in the 1990’s. Journal of Occupational Medicine 1990; 32:1199-1206.
3 Duff LA, Kitson AL, Seers K, Humphris D. Clinical guidelines: An introduction to their development and implementation. Journal of Advanced Nursing 1996; 23: 887-895.
4 Hutchison A. The philosophy of clinical practice guidelines: purposes, problems, practicality and implementation. Journal of Quality Clinical Practice 1998; 18: 63-73.
5 Mead P. Clinical guidelines: promoting clinical effectiveness or a professional minefield? Journal of Advanced Nursing 2000; 31:110-116.
6 Ouchi WG. The transmission of control through organisational hierarchy. Academy of Management Journal 1978; 12: 173-192.
7 Westrum R.. Cultures and requisite imagination. In: Wise J, Hopkin D, editors. Verification and validation: Human Factors aspects. New York:Springer;1993. p.401-416.
8 Thomas LH, McColl E, Cullum N, Rousseau N, Soutter J, Steen N. Effect of clinical guidelines in nursing, midwifery and the therapies: a systematic review of evaluations. Quality in Health Care 1998; 7:183-191.
9 Grimshaw J, Freemantle N, Wallace S, Russell I, Hurwitz B, Watt I et al. Developing and implementing clinical practice guidelines. Quality in Health Care 1995; 4: 55-64.
10 Getting evidence into practice. Effective Health Care Bulletin 1999; 5: 1.
11 Honingsbaum F, Ham C. Improving Clinical Effectiveness: The development of clinical guidelines in the West Midlands. Birmingham: Health Services Management Centre;1996.
12 Perrow C. A framework for the comparative analysis of organisations. American Sociological Review 1967;32:194-208.
13 Hale AR, Swuste P. Safety rules: procedural freedom or action constraint. Safety Science 1998; 29:163-177.
14 Kaempf GL, Orasnu J. Current and future applications of naturalistic decision making in aviation. In: Zsambok C, Klein G, editors. Naturalistic Decision Making New Jersey:Lawrence Erlbaum; 1997. p.81-90.
15 Lawton R, Parker D. Procedures and the professional: the case of the British NHS. Social Science and Medicine 1999; 48: 353-361.
16 Department of Health, Chapter 2. In: National Service Framework for Coronary Heart Disease. Modern Standards and Service Models, London: Stationary Office 2000
17 Reason J, Parker D, Lawton R. Organizational controls and safety: The varieties of rule-related behaviour. Journal of Occupational and Organizational Psychology 1998; 71: 289-304.
18 Lloyd B. Guidelines to reduce risk and improve the quality and cost-effectiveness of care. Managing Risk 1995: April; 5-6.
19 Grilli R, Lomas J. Evaluating the message: the relationship between compliance rate and the subject of a practice guideline Medical Care 1994; 32: 202-13.
20 Cushing S. Fatal Words:Communication Clashes and Aircraft Crashes. Chicago: University of Chicago Press; 1994.
21 Horning LA, Smith PW. Infection Control Violations. Infection Control and Hospital Epidemiology 1991;12: 672-675.
22 Lawton R. Not working to rule: understanding procedural violations at work. Safety Science 1998; 28: 77-95
23 Grol R. National standard setting for quality of care in general practice: attitudes of general practitioners and response to a set of standards. British Journal of General Practice 1990; 40: 361-364.
24 Tunis SR, Hayward RSA, Wilson MC, Rubin HR, Bass EB, Johnston M et al. Internists’ attitudes about clinical practice guidelines. Annals of Internal Medicine 1994; 120:956-963.
25 Hewison J, Wilson, B. Policies, plans and individual behaviour change in the implementation of evidence based practices. BPS Annual Conference, Health Psychology Section; 1999 Sep1-3; University of Leeds.
|HOME||HELP||FEEDBACK||BROWSE ARTICLES||BROWSE BY AUTHOR|