Patients’ failure to follow the advice of health professionals is known as non-compliance and is generally considered to be a substantial problem.
VOL: 102, ISSUE: 02, PAGE NO: 28
Sherrill Snelgrove, MPhil, BSc, RGN, is a lecturer at the School of Health Science, University of Wales, Swansea
It is seen as costly for patients in terms of persistence of disease or worsening health and for the NHS in terms of wasted resources. Non-compliance can also damage the relationship between health professional and patient and may even lead to wider costs. Patients’ failure to complete courses of antibiotics, for example, is often cited as a factor in the increased resistance of many bacteria.
Walker et al (2004), however, point out that compliance can also be problematic, for example if patients continue to take medication unquestioningly, even though it is causing adverse side-effects.
Compliance, adherence, concordance
The word ‘compliance’ itself is problematic. It is often used interchangeably with ‘adherence’ or ‘concordance’, leading to a lack of clarity, and implies unquestioning obedience with no opportunity for choice. Haynes (1979) described compliance as ‘the extent to which a person’s behaviour (taking medication, following a recommended diet or executing lifestyle changes) coincides with medical or health advice’. This medical definition has been heavily criticised in nursing literature for its paternalistic view of the health professional-patient relationship in which the patient is perceived as ‘passive’ and expected to obey the clinician’s orders. By not conforming to clinicians’ expectations, patients may be viewed as deviant and labelled non-compliant. Many nurses feel uneasy about the label as it directs blame largely towards the patient (Russell et al, 2003).
This picture of health professional and patient interaction is at odds with more recent discourse, which sees the patient as being an active participant in decisions about treatments. Annandale and Hunt (1998) suggest that many patients do evaluate treatments and base their actions on informed decisions. This is demonstrated by the recent lack of uptake of the measles, mumps and rubella (MMR) vaccine - Bellaby (2003) suggests parents weigh up the available evidence and act in what they see as their children’s best interests rather than complying unquestioningly with medical advice.
Other terms have been introduced to avoid patients being viewed as either passive or non-compliant. ‘Adherence’ has been used but while it emphasises negotiation between clinician and patient, it still implies a degree of passivity and obedience.
More recently ‘concordance’ has been promoted because it implies the patient is an equal partner, supporting the ethos of shared decision-making between patient and health professional rather than more traditional paternalism (Weiss and Britten, 2003).
Measuring poor concordance
Compliance can be measured by objective and non-objective means. Non-objective methods include asking patients whether they have taken the prescribed medication, pill or bottle counts and assessing therapeutic outcomes. However, these methods are fraught with problems as patients may not admit non-compliance, forget whether they have complied over a long period, take the correct medication but on the wrong schedule or simply recover without actually complying. Objective methods such as blood, urine or serum assays are less prone to bias than non-objective measures but are relatively expensive, do not account for individual metabolism and are simply snapshots that do not reflect inconsistencies in compliance over time.
The scale of the problem
While measuring compliance can be difficult, there is strong evidence that it is a widespread problem. Up to 80 per cent of patients can be expected not to comply with their treatment at some time (Dunbar-Jacob et al, 1995). Non-compliance affects nearly all disease areas (Medicines Partnership, 2005) and involves an estimated 38 per cent of patients on short-term treatment and 43 per cent on long-term treatment (DiMatteo, 1994).
One early study revealed that only 12 per cent of patients who were prescribed penicillin for 10 days were still taking it by the tenth day; 50 per cent had stopped by the third day, 71 per cent by the sixth and 82 per cent by the ninth (Bergman and Werner, 1963).
Non-compliance can have serious outcomes. Christensen and Moran (1998) estimated that it accounted for approximately 50 per cent of deaths in patients with renal failure and the majority of renal transplant failures in the second year. Bandolier (2005) reported a study that found patients taking statins had stopped taking the drug or took less than prescribed within a year. This reduces or removes any benefit and may even cause harm.
Reasons for poor concordance
There are many reasons for poor concordance (Box 1). Research has focused on:
- Communication processes;
- Health beliefs;
- Social context;
- Intrapsychic processes.
Ley (1989) proposed that patient compliance can be predicted by the patient’s degree of satisfaction with the consultation, level of understanding and accuracy of recall of information, and suggests it may be improved by:
- Bearing in mind that patients tend to remember the first part of the message;
- Stressing the importance of compliance;
- Simplifying information given;
- Using repetition;
- Being specific;
- Following up with additional interviews.
Ley’s model (1989) is supported by a number of studies (Ogden, 2004) and acknowledges the importance of effective communication and its relationship to compliance in health care. However, it is an educational model based on the assumption that the clinician is the objective, authoritative educator and expert and the patient the naive learner. Ley does not address the influence of patients’ health beliefs or the social context in which they are operating. Stanton (1987) responded to these issues and developed a model of adherence that emphasises wider issues that may affect compliance:
- The importance of partnership between clinician and patient;
- Patients’ social support and circumstances, health beliefs and beliefs about lifestyle disruption;
- Patients’ sense of control (locus of control).
Lay beliefs, or non-medical beliefs, can be a powerful source of non-compliant behaviour (Meichenbaum and Turk,1987). Health professionals may not see lay beliefs as having the same validity as medical information but understanding patients’ health beliefs will help clinicians to understand poor concordance.
From this perspective, compliance is related to the patient’s view of illness and treatment rather than the clinician’s view (Janz and Becker, 1984). Traditionally, clinicians see themselves as being informed and as being consulted by uninformed patients. However, this may be inconsistent with patients’ views. They may hold beliefs about their symptoms that derive from a variety of sources and compare the clinician’s decisions with these. If they compare unfavourably, the result may be lack of concordance.
For example, patients with chronic illnesses may feel a sense of authority over general clinicians as they have a more intimate knowledge of their condition. Indeed, those with chronic illness are often better informed than the clinicians who look after them. They will weigh up the options and, if the advice they are given conflicts with their own beliefs, poor concordance may ensue.
Patients’ beliefs about treatment may help clinicians to understand poor concordance. For example, drugs may be seen as toxic and ‘drug holidays’ as helping to reduce any toxicity. Alternatively consistent long-term use of particular drugs may lead to addiction and make patients reluctant to change drug regimens.
Social context and cultural differences
Russell et al (2003) suggest that nurses need to consider a social model that puts greater emphasis on recognising patients’ social environments (such as the amount of social support available) or physical environment (a young mother may not have funds or babysitters to enable her to attend clinics). While there may be a risk of ‘over-surveillance’, Russell et al (2003) argue that this information should be seen as health-related in order to understand why patients’ circumstances may influence non-compliance.
Studies have shown that where nurses consider patients’ social circumstances and individuality rather than only the disease processes, compliance is higher. Kyngas (2000) reported that support from nurses for adolescents with asthma was the most important factor in predicting compliance. Similarly, McCann and Clark (2003) found young adults with schizophrenia more likely to attend mental health services when they had good relationships with, and ‘open access’ to, nurses.
Nurses and patients come from a variety of cultural backgrounds and have their own health beliefs, which will influence their interactions. Good nursing care incorporates an understanding of how these beliefs affect both nursing practice and patients’ concordance with treatment regimens. Delivering information and advice to sick and distressed patients can be extremely difficult, and efficacy is often reduced when nurse and patients are from different ethnic backgrounds. Language difficulty is one of the main barriers to effective communication and subsequent concordance but there may also be misunderstanding or ignorance about the patient’s health beliefs and practices. Central to ethnocentric nursing is effective communication, taking into account ‘appropriate cultural knowledge and an empathetic attitude towards minority ethnic clients (Gerrish et al, 1996).
Intentional non-compliance may be patients’ means of exerting control over their lives. Kingsnorth and Wilkinson (1996) investigated non-compliance in discharged patients receiving palliative care, and found that over half the reasons given were that the patients believed they did not need the medication or regulated themselves and took it when required. Both illness and treatment may be perceived as disempowering, and patients may react to a lack of autonomy with non-compliance.
Nurse management of concordance
Nurses have a long history of initiating effective interventions that have resulted in improved concordance in many aspects of health care regimens. For example, improving the uptake of cervical screening by refining cytology tracking systems; suggesting improved health education services for women from ethnic groups (Elsworth, 2001); improving access to mental health clinics for young adults with schizophrenia (McCann and Clark, 2003); and supporting and educating hypertensive patients (Morris et al, 1997).
However, in conjunction with interventions for improving concordance (Box 2), nurses are in a prime position to support and develop a model of communication where patients’ beliefs are acknowledged and they are asked the extent to which they want to play a part in decision-making about their own health (Weiss and Britten, 2003).
To this end nurses can draw on their own clinical experience but may also wish to develop academic knowledge about memory processes and information-giving, health beliefs, intrapsychic processes and education in health care.
Each week Nursing Times publishes a guided learning article with reflection points to help you with your CPD. After reading the article you should be able to:
- Understand what is meant by concordance and lack of concordance;
- Identify types of poor concordance;
- Be familiar with methods of assessing lack of concordance;
- Understand the reasons for poor concordance;
- Be able to identify strategies for improving patients’ concordance.
Use the following points to write a reflection for you PREP portfolio
- Write about where you work and why this article is relevant;
- Reflect on the last time you encountered a concordance issue with a patient;
- Identify any information in this article that you could have used in that situation;
- Write about how you will use this knowledge in your future practice;
- Summarise what you will do to follow up what you have learnt.
This article has been double-blind peer-reviewed.
For related articles on this subject and links to relevant websites see www.nursingtimes.net