- Email: [email protected]
Medication adherence: predictive factors and enhancement strategies
Specific community treatments
Medication adherence: predictive factors and enhancement strategies
What’s new? • Predictors of non-adherence and adherence enhancing strategies have been more extensively researched for bipolar disorder
Maxine X Patel
• Prescription refill data has been used in a large-scale study in schizophrenia, which found that 59% were not fully adherent
Anthony S David
• Financial incentives for medication adherence, previously used for substance abuse disorders and tuberculosis, are now subject to research studies for psychiatric disorders
Abstract Medication adherence is crucial in psychiatry, especially for chronic disorders. Both clinician and patient share responsibility for adherence, which is rarely an all-or-none phenomenon. For psychiatric drugs, non-adherence rates are approximately 40–60%. Such non-adherence explains much of the difference between drug efficacy and effectiveness, as demonstrated by higher relapse rates in non-adherent patient groups. Thus, non-adherence impacts profoundly on clinical and economic burdens for health services. During clinical assessment, predictive factors of non-adherence should be considered, including: a prior history of nonadherence; alcohol or substance misuse or where treatment is during an asymptomatic phase. Similarly, drug dose and formulation polypharmacy, side effects and the therapeutic relationship also affect adherent behaviour. Psychoeducational interventions aiming to enhance adherence focus primarily on imparting knowledge, rather than on attitudinal and behavioural change, and have proved largely ineffective. Individual psychological interventions are more effective as they specifically target the patient’s beliefs and attitudes concerning the illness and medication by utilizing cognitive–behavioural or motivational interviewing techniques. Compliance therapy combines all of these. All clinicians should routinely use simple adherence-enhancing techniques, particularly as dedicated resources for specialist interventions remain rare. Moreover, in an attempt to further reduce the adverse clinical and economic impact of non-adherence, it is imperative that patients are given the opportunity to have their personal individual perspectives adequately heard.
• New legal measures (e.g. ‘supervised community treatment’) can require compulsory medication adherence in the community
Adherence to medication is crucial in psychiatry, as in other medical specialties, especially those focusing on chronic dis orders. This contribution outlines the associated factors and considers enhancement strategies.
Terminology and definitions Compliance is generally defined as the extent to which a per son’s behaviour coincides with medical advice. This implies that it is the patient’s responsibility to follow the clinician’s orders and thus has been criticized as paternalistic. Adherence includes the concepts of patient choice: both clini cian and patient share the responsibility for adherence. Thus, the prescriber should give understandable instructions regarding medication so that the patient can then interpret them correctly.
Keywords adherence; compliance therapy; health belief model; interventions; measurement tools; predictive factors
Concordance is based on the notion that the therapeutic alliance between the prescriber and patient is a negotiation process, with equal respect for both the patient’s and clinician’s agenda.1–3 There is no current universally accepted definition for medi cation adherence. In most research, definitions for adherence are usually dichotomous but adherence is rarely an all-or-none phenomenon and can include: • intentional or accidental errors in dosage and timing • total or partial omission • use of inadvertent combinations.4–6 Measuring adherence is not straightforward and various methods exist, not least of all because clinicians significantly underestimate rates of non-adherence in their patient populations (see Table 1).3,5–8 One study used prescription refill data to measure adherence rates for antipsychotics in schizophrenia and found that only 41% of patients were adherent (i.e. ‘cashed’ their prescription regularly), 16% were partially adherent, 24% were non-adherent and 19% were excess fillers.9
Maxine X Patel MSc MRCPsych is a Clinical Lecturer at the Institute of Psychiatry, King’s College London, UK, and a Consultant at the South London and Maudsley NHS Foundation Trust. Her research interests include pharmacoepidemiology, medication adherence and attitudes to antipsychotic drugs. Conflicts of interest: she was funded by a special training fellowship from the Medical Research Council. She has previously worked on two clinical drug trials for Janssen-Cilag and has also been reimbursed for attendance costs at scientific conferences and received consultation and speaker fees from the pharmaceutical industry. Anthony S David MD FRCPsych is Professor of Cognitive Neuropsychiatry at the Institute of Psychiatry, King’s College London, UK, and Honorary Consultant at the South London and Maudsley NHS Foundation Trust. His research interests include insight into illness and neuropsychological aspects of psychiatric disorders. Conflicts of interest: he has worked on two clinical drug trials for Janssen-Cilag and has received consultation and speaker fees from the pharmaceutical industry.
PSYCHIATRY 6:9
357
© 2007 Elsevier Ltd. All rights reserved.
Specific community treatments
However, it should be remembered that the association between non-adherence and relapse is bi-directional, as nonadherence can lead to symptom relapse and deteriorating symptoms can precipitate erratic adherence.4,5 In essence, nonadherence has a profound impact on the clinical and economic burden for health services.15,16
Adherence measurement tools Self-report methods These include interviews and questionnaires (e.g. Tablet Routines Questionnaire) which assess the daily routines for taking medication and the proportion of medication an individual has missed in the previous week and previous month5
General factors and adherence-enhancing strategies
Pill counts These are time-consuming, labour intensive and there is great potential for inaccuracy. Adherence (%) can be calculated as (the number of pills taken / the number of pills prescribed) × 1007
Many different factors affect adherence and these vary between patients. Consideration of these factors during a clinical assess ment should be the first stage of any intervention to enhance adherence for a patient (see Table 2).
Electronic methods Electronic devices have been developed which can be attached to the tablet bottle. They record the time and date on every occasion that the bottle is opened
Patient and illness characteristics Sociodemographics: systematic reviews of the literature high light that age at illness onset, age at first hospitalization, sex, socioeconomic status, marital status and ethnicity are not consis tently associated with differences in medication adherence.16,17
Prescription (refill) monitoring The frequency of prescription dispensing for an individual can be monitored as a proxy measure of adherence
Adherence history: partially adherent patients are more likely to have a past history of non-adherence than fully adherent patients.5 Thus, early detection and management of nonadherence may be of particular importance.
Saliva, plasma and urine assay tests These are the most objective measure of adherence but do not exist for all psychiatric drugs and are expensive and more invasive. Furthermore, assay tests have limited value in assessing partial adherence and may overestimate adherence with drugs that have a long half-life3,6,8
Illness duration: in chronic diseases, adherence usually declines over time and is generally lower when the patient is asymptom atic, treatment is prophylactic and the consequences of stopping treatment are delayed.2,8 However, in patients with first-episode psychosis, non-adherence rates can be high (40%) and beliefs about need for medication appear particularly important.
Table 1
Psychopathology: hostility, suspiciousness, disorganization and ‘hypomanic denial’ have been shown to be significantly associ ated with non-adherence in psychotic illnesses. However, global measures of symptom severity in schizophrenia do not appear to be consistent predictors of adherence.14,16
Magnitude and impact of non-adherence Non-adherence rates are reported as 40–60% for antipsychotics, 18–56% for mood stabilizers and 30–97% (median 63%) for antidepressants.5,6,10 Whilst these estimates might seem unac ceptably high, similar (if slightly lower) rates are seen in all other medical specialities. Medication non-adherence explains much of the difference between drug efficacy under ideal clinical circumstances and drug effectiveness seen in routine clinical practice.
General strategies for adherence enhancement
Relapse: non-adherent patients with schizophrenia are 3.5 times more likely than adherent patients to relapse within two years.8
• Questions about adherence should be routinely included in clinical assessment and repeated during follow-up, as adherence declines over time • Be alert to the impact of cognitive deficits, alcohol and illicit substance use on adherence • Simplify dosing frequency and polypharmacy where possible, and consider potential benefits of alternative drug formulations • Consider potential detrimental effect of frequency and severity of side effects on adherence. Titrate to minimal optimal dose for good efficacy but few side effects • Ensure instructions about medication are adequately understood by the patient and enlist family support where appropriate
Rehospitalization: non-adherence accounts for 40% of re hospitalization costs within 2 years.11 Time to remission: repeated relapse, often due to non-adherence, is associated with an increased difficulty or delay in achieving remission.12 Clinical burden and cost: non-adherence has a predicted excess annual cost of £2500 per patient for inpatient services and over £5000 for total service use.13 Attempted suicide: in both schizophrenia and bipolar disorder, risk of suicide attempt has been found to be significantly higher in patients not adhering to their medication regimens.12,14
PSYCHIATRY 6:9
Table 2
358
© 2007 Elsevier Ltd. All rights reserved.
Specific community treatments
Insight: illness beliefs are important predictors of adherence, although patients with no insight into illness may still accept and derive benefit from medication. Thus, helping patients to accept a mental illness label is less important than enhancing awareness of the impact of medication on their daily functional ability.2,16,17
Other significant factors Clinician: a poor clinician–patient relationship will adversely affect adherence. Clinician factors associated with adherence include: • the extent to which the clinician values medication • job satisfaction, i.e. morale and enthusiasm • continuity of clinician.4,8,16,18
Cognitive deficits and use of prompts: the patient’s ability to understand the purpose of medication and follow his or her medication regimen may be limited by cognitive deficits. Hence, cognitive and memory-enhancing strategies (e.g. external prompts such as telephone reminders) may be used to enhance adherence.2,8
Coercion: within the framework of the Mental Health Act 1983, psychiatrists can enforce medication against a patient’s will. Non-adherence is associated with the use of such measures, par ticularly with regard to involuntary admission. Thereafter, some patients make a smooth transition to voluntary treatment having directly experienced the benefits, whereas others remain resent ful. It can therefore be hard to establish a therapeutic alliance. Recent changes in law have created new orders called ‘super vised community treatment’ (England and Wales) and ‘compul sory treatment orders’ (Scotland), which can require compulsory medication adherence in the community. How effective these will be remains to be fully evaluated.
Comorbidity: alcohol and substance misuse are strong predic tors of non-adherence. Sometimes patients quote their clinician’s advice not to mix medication with alcohol as a reason for nonadherence.8,14,16 Medication and side effects Dose strength: the relationship between dose strength and adherence remains unclear but is probably curvilinear, with very low doses being associated with poor efficacy and very high doses with excessive side effects.8,16 High doses are sometimes prescribed because of apparent lack of pharma cological effectiveness, when the real cause is in fact partial adherence.
Clinic setting and organization: improvements to décor and ambience, offering refreshments and maintaining a realistic appointment schedule are all claimed to be associated with improved adherence.1,8 Access: lack of access to care and practical barriers (e.g. finan cial) have been highlighted, particularly for homeless persons, who are also often viewed as non-adherent. Hence clinicians should assist eligible patients in claiming for free prescriptions.7
Dosing frequency: it has been suggested that once-daily dosing is better than three times per day but that once-daily dosing is more frequently associated with partial adherence.4
Financial incentives (money, cash or vouchers redeemable for other goods) have been found to enhance adherence in substance abuse and non-psychiatric disorders, and are more cost-effective than other strategies. However, there are ethical concerns regarding the use of financial incentives for patients with major psychiatric disorders.
Drug formulation and route of administration: depot injections are associated with higher adherence rates than oral formula tions, drugs in liquid form higher rates than tablets, and sublin gual higher rates than normal tablets. Depot antipsychotics may be associated with lower relapse rates than antipsychotic tablets and afford early detection of non-adherence, thereby enabling early intervention. However, merely switching patients to depot without considering their reasons for oral non-adherence is unwise.4,8
Family support is associated with better adherence, although stressful social interactions and shared superstitions or false beliefs about medications may negate this effect. Generally, it is advoca ted that family support be enlisted. Alternatively, formal super vised medication administration – by members of the community mental health team (CMHT), for example – can be arranged.7
Drug packaging: calendar packs for tablets have been associ ated with better adherence than normal packaging (e.g. tablets in strips of ten).4 With a complex drug regimen, appropriate packaging may enable the patient to take the correct dose at the correct time. Alternatively, a ‘dosette box’ may be used.
Community-based interventions: components of community programmes have included provision of a strong and support ive social network and stable housing, other supportive services and close monitoring of clinical status. These can be effective in improving adherence in patients with psychosis.6,19
Polypharmacy: minimizing non-essential polypharmacy or com bination therapy (e.g. unnecessary anticholinergics or hypnotics) can improve intentional and unintentional non-adherence.
Psychological factors and adherence-enhancing interventions Factors underpinning non-adherent behaviour include the patient’s beliefs and attitudes concerning the illness and medication. Various psychological interventions specifically target these to enhance adherence (see Table 3).
Side effects: medication side effects are commonly cited by patients as their main reason for non-adherence, perhaps because clinicians consistently underestimate both the frequency and severity of side effects. As side effects do not appear to be a consistent predictor of non-adherence, it has been suggested that it is the fear rather than the actual experience of side effects that increases the risk of non-adherence.5,16
PSYCHIATRY 6:9
The health belief model The health belief model outlines four main belief categories: ben efits; costs; susceptibility; and secondary benefits of medication 359
© 2007 Elsevier Ltd. All rights reserved.
Specific community treatments
Family and group: family-directed interventions assume that the family unit influences the course of a patient’s illness. Such programmes usually include psychoeducation and are, in them selves, largely ineffective in enhancing adherence for schizophre nia, but strategies which involve the family are deemed more effective in bipolar disorder.14 Merely imparting knowledge that is not specific and personal to the individual does not necessar ily lead to a change in behaviour. Group interventions are based on the importance of peer support and shared identification and are usually based on psychoeducational principles. Again, these have largely been shown to be ineffective unless they also focus on behavioural change.6
Psychological interventions for adherence enhancement • Aim to understand the patient’s personal construct of the meaning of medication and illness • Consider the patient’s individual cost–benefit analysis of medication in the context of their personal goals and priorities • Be willing to acknowledge the potential positive value of symptoms and negative consequences of taking medication • Remember that simple psychoeducational programmes generally do not improve adherence • Consider using behavioural modification strategies (e.g. reminders)
Individual psychological interventions Cognitive–behavioural therapy (CBT) comprises two main com ponents and has been seen to be effective for adherence enhance ment in bipolar disorder.14 The cognitive component effectively targets the patient’s attitudes and beliefs towards medication and assumes that adherence is determined by the personal construc tion of the meaning of medication and illness. The behavioural component is based on the assumption that behaviour is learnt and can be modified, and so patients are provided with instruc tions and strategies (e.g. reminders, self-monitoring tools, cues and reinforcements); these can also improve adherence.6
Table 3
and adherence. It focuses on the patient’s decision-making pro cess, emphasizing the subjective cost–benefit analysis of any treatment, including probabilities of advantages and likely bur dens, in the context of the patient’s personal goals and priori ties. The underlying premise is that adherence enhancement is possible if the patient’s perceptions alter (see Table 4). However, this model may have limited predictive value for non-adherent behaviour. Application of the model to adherence-enhancing strategies can be compromised by a patient’s impaired cognitive function, limiting their capacity in risk assessment and planning. Moreover, a patient’s attitude to medication may be completely different from their actual medication-taking behaviour.2,6,8
Motivational interviewing enables the patient to express per sonal reasons for and against improving their adherence behavi our. Importantly, the clinician is required to acknowledge the potential positive value of symptoms and also the negative consequences of taking medication before the patient may even tually decide that symptom control by medication is preferable.2
Psychoeducational interventions Individual: many clinicians believe that patient education regarding illness, relapse prevention and medication will directly improve adherence. However, as most psychoeducational pro grammes focus primarily on imparting knowledge rather than on attitudinal and behavioural change, these have proved largely ineffective in enhancing adherence, although they remain more popular for bipolar disorder. Furthermore, increasing knowledge about illness and medication may be disturbing to the patient, as it may be a direct challenge to their only partial insight, although the patient’s fundamental right to information should not be compromised.6,8,14,17,18
Compliance therapy is a brief adherence-enhancing intervention based on motivational interviewing and cognitive approaches. Patients express their beliefs and concerns about medication and focus on personal goals as well as the more indirect benefits of medication, including improved personal relationships. Ambi valence towards medication is explored, consequences of medi cation cessation are discussed, analogies with chronic physical illness are made, and the pros and cons of medication are con sidered. The aim is to highlight the role of medication in staying well to fulfil needs and achieve lifestyle and personal goals. Com pliance therapy is advantageous in terms of insight, attitudes to treatment, improved social functioning and adherence, immedi ately post-therapy as well as at 18-month follow-up.3 However, in a more recent multicentre, European randomized controlled trial (RCT) comparing compliance therapy with a health edu cation intervention, participants in both groups improved. The study failed to show a benefit for compliance therapy over the control intervention, possibly because the control intervention also improved adherence.20
The health belief model Four main belief categories: • benefits • costs • susceptibility • secondary benefits of medication and adherence Adherence enhancement is improved if patients believe: • they are more likely to experience benefits from adherence • there will be few personal costs resulting from adherence • they will be at high risk of becoming seriously unwell if they do not adhere
Research limitations • Many adherence studies are subject to selection bias of re search participants and are also limited in the generalizability of their findings. Non-adherent patients are, by definition, more difficult to recruit to such studies, for the same reasons that they are non-adherent in the first place, such as homelessness.
Table 4
PSYCHIATRY 6:9
360
© 2007 Elsevier Ltd. All rights reserved.
Specific community treatments
• Retrospective study designs are commonplace, but accurate calculation of adherence rates and true identification of risk factors for non-adherence warrants prospective studies. • Most studies have concentrated on adherence only in psych oses or mood disorders. • There are numerous methods for measuring medication ad herence, with differing criterion thresholds used for defining non-adherence; there is thus little agreement between methods.15 Many methods do not truly measure if the patient actually in gested the medication and no single method is ideal. • Studies on adherence enhancement often involve complex in terventions in which the specific active component is not identi fied. They can be labour intensive and so may not readily be applied in everyday clinical practice. Studies often do not have a sufficient duration of follow-up to measure enduring effective ness, and measure adherence rates only, not clinical outcomes such as reduction in psychopathology.6,18
4 Demyttenaere K. Compliance during treatment with antidepressants. J Affect Disord 1997; 43: 27–39. 5 Scott J, Pope M. Nonadherence with mood stabilizers: prevalence and predictors. J Clin Psychiatry 2002; 63: 384–90. 6 Zygmunt A, Olfson M, Boyer CA, Mechanic D. Interventions to improve medication adherence in schizophrenia. Am J Psychiatry 2002; 159: 1653–64. 7 Azrin NH, Teichner G. Evaluation of an instructional program for improving medication compliance for chronically mentally ill patients. Behav Res Ther 1998; 36: 849–61. 8 Fenton WS, Blyler CR, Heinssen RK. Determinants of medication compliance in schizophrenia: empirical and clinical findings. Schizophr Bull 1997; 23: 637–51. 9 Gilmer TP, Dolder CR, Lacro JP, et al. Adherence to treatment with antipsychotic medication and health care costs among Medicaid beneficiaries with schizophrenia. Am J Psychiatry 2004; 161: 692–99. 10 Pampallona S, Bollini P, Tibaldi G, Kupelnick B, Munnizza C. Patient adherence in the treatment of depression. Br J Psychiatry 2002; 180: 104–09. 11 Weiden PJ, Olfson M. Cost of relapse in schizophrenia. Schizophr Bull 1995; 21: 419–29. 12 Leucht S, Heres S. Epidemiology, clinical consequences, and psychosocial treatment of nonadherence in schizophrenia. J Clin Psychiatry 2006; 67(suppl 5): 3–8. 13 Knapp M, King D, Pugner K, Lapuerta P. Non-adherence to antipsychotic medication regimens: associations with resource use and costs. Br J Psychiatry 2004; 184: 509–16. 14 Sajatovic M, Davies M, Hrouda DR. Enhancement of treatment adherence among patients with bipolar disorder. Psychiatr Serv 2004; 55: 264–69. 15 Dolder CR, Lacro JP, Leckband S, Jeste DV. Interventions to improve antipsychotic medication adherence: review of the recent literature. J Clin Psychopharmacol 2003; 23: 389–99. 16 Lacro JP, Dunn LB, Dolder CR, Leckband SG, Jeste DV. Prevalence of and risk factors for medication nonadherence in patients with schizophrenia: a comprehensive review of recent literature. J Clin Psychiatry 2002; 63: 892–907. 17 Colom F, Vieta E, Tacchi MJ, Sanchez-Moreno J, Scott J. Identifying and improving non-adherence in bipolar disorders. Bipolar Disord 2005; 7(suppl 5): 24–31. 18 Haynes RB, McDonald H, Garg AX, Montague P. Interventions for helping patients to follow prescriptions for medications (Cochrane Review). The Cochrane Library, Issue 3. Oxford: Update Software, 2003. 19 Nosé M, Barbui C, Gray R, Tansella M. Clinical interventions for treatment non-adherence in psychosis: meta-analysis. Br J Psychiatry 2003; 183: 197–206. 20 Gray R, Leese M, Bindman J, et al. Adherence therapy for people with schizophrenia: European multicentre randomised controlled trial. Br J Psychiatry 2006; 189: 508–14.
Conclusion Research which has attempted to identify general factors that predict variance in adherence has generated mostly contradic tory findings. However, psychological research attempting to understand why patients are not adherent has found attitudes and behaviours to be better predictors of non-adherence.4,8 Simple psychoeducational interventions are currently less successful in enhancing adherence than might be predicted and cognitive–behavioural techniques are more likely to succeed. However, dedicated time and resources for such interventions remain rare, despite high rates of non-adherence being common place. Such psychological therapies should not stand alone and employing a combination of strategies may be more favourable.15 Adherence-enhancing techniques should be routine for all clini cians, supported by the multidisciplinary team. Adherence studies underline the need to strive to understand our patients from their personal individual perspectives. This means giving them the opportunity to have their voices adequately heard. In so doing, it is hoped that the adverse clinical and eco nomic impact of non-adherence may be further reduced. ◆
References 1 Corrigan PW, Liberman RP, Engel JD. From non-compliance to collaboration in the treatment of schizophrenia. Hosp Community Psychiatry 1990; 41: 1203–11. 2 Hughes I, Hill B, Budd R. Compliance with antipsychotic medication: from theory to practice. J Ment Health 1997; 6: 473–89. 3 Kemp R, Kirov B, Hayward P, David A. Randomised control trial of compliance therapy. Br J Psychiatry 1998; 172: 413–19.
PSYCHIATRY 6:9
361
© 2007 Elsevier Ltd. All rights reserved.
Medication adherence: predictive factors and enhancement strategies
What’s new? • Predictors of non-adherence and adherence enhancing strategies have been more extensively researched for bipolar disorder
Maxine X Patel
• Prescription refill data has been used in a large-scale study in schizophrenia, which found that 59% were not fully adherent
Anthony S David
• Financial incentives for medication adherence, previously used for substance abuse disorders and tuberculosis, are now subject to research studies for psychiatric disorders
Abstract Medication adherence is crucial in psychiatry, especially for chronic disorders. Both clinician and patient share responsibility for adherence, which is rarely an all-or-none phenomenon. For psychiatric drugs, non-adherence rates are approximately 40–60%. Such non-adherence explains much of the difference between drug efficacy and effectiveness, as demonstrated by higher relapse rates in non-adherent patient groups. Thus, non-adherence impacts profoundly on clinical and economic burdens for health services. During clinical assessment, predictive factors of non-adherence should be considered, including: a prior history of nonadherence; alcohol or substance misuse or where treatment is during an asymptomatic phase. Similarly, drug dose and formulation polypharmacy, side effects and the therapeutic relationship also affect adherent behaviour. Psychoeducational interventions aiming to enhance adherence focus primarily on imparting knowledge, rather than on attitudinal and behavioural change, and have proved largely ineffective. Individual psychological interventions are more effective as they specifically target the patient’s beliefs and attitudes concerning the illness and medication by utilizing cognitive–behavioural or motivational interviewing techniques. Compliance therapy combines all of these. All clinicians should routinely use simple adherence-enhancing techniques, particularly as dedicated resources for specialist interventions remain rare. Moreover, in an attempt to further reduce the adverse clinical and economic impact of non-adherence, it is imperative that patients are given the opportunity to have their personal individual perspectives adequately heard.
• New legal measures (e.g. ‘supervised community treatment’) can require compulsory medication adherence in the community
Adherence to medication is crucial in psychiatry, as in other medical specialties, especially those focusing on chronic dis orders. This contribution outlines the associated factors and considers enhancement strategies.
Terminology and definitions Compliance is generally defined as the extent to which a per son’s behaviour coincides with medical advice. This implies that it is the patient’s responsibility to follow the clinician’s orders and thus has been criticized as paternalistic. Adherence includes the concepts of patient choice: both clini cian and patient share the responsibility for adherence. Thus, the prescriber should give understandable instructions regarding medication so that the patient can then interpret them correctly.
Keywords adherence; compliance therapy; health belief model; interventions; measurement tools; predictive factors
Concordance is based on the notion that the therapeutic alliance between the prescriber and patient is a negotiation process, with equal respect for both the patient’s and clinician’s agenda.1–3 There is no current universally accepted definition for medi cation adherence. In most research, definitions for adherence are usually dichotomous but adherence is rarely an all-or-none phenomenon and can include: • intentional or accidental errors in dosage and timing • total or partial omission • use of inadvertent combinations.4–6 Measuring adherence is not straightforward and various methods exist, not least of all because clinicians significantly underestimate rates of non-adherence in their patient populations (see Table 1).3,5–8 One study used prescription refill data to measure adherence rates for antipsychotics in schizophrenia and found that only 41% of patients were adherent (i.e. ‘cashed’ their prescription regularly), 16% were partially adherent, 24% were non-adherent and 19% were excess fillers.9
Maxine X Patel MSc MRCPsych is a Clinical Lecturer at the Institute of Psychiatry, King’s College London, UK, and a Consultant at the South London and Maudsley NHS Foundation Trust. Her research interests include pharmacoepidemiology, medication adherence and attitudes to antipsychotic drugs. Conflicts of interest: she was funded by a special training fellowship from the Medical Research Council. She has previously worked on two clinical drug trials for Janssen-Cilag and has also been reimbursed for attendance costs at scientific conferences and received consultation and speaker fees from the pharmaceutical industry. Anthony S David MD FRCPsych is Professor of Cognitive Neuropsychiatry at the Institute of Psychiatry, King’s College London, UK, and Honorary Consultant at the South London and Maudsley NHS Foundation Trust. His research interests include insight into illness and neuropsychological aspects of psychiatric disorders. Conflicts of interest: he has worked on two clinical drug trials for Janssen-Cilag and has received consultation and speaker fees from the pharmaceutical industry.
PSYCHIATRY 6:9
357
© 2007 Elsevier Ltd. All rights reserved.
Specific community treatments
However, it should be remembered that the association between non-adherence and relapse is bi-directional, as nonadherence can lead to symptom relapse and deteriorating symptoms can precipitate erratic adherence.4,5 In essence, nonadherence has a profound impact on the clinical and economic burden for health services.15,16
Adherence measurement tools Self-report methods These include interviews and questionnaires (e.g. Tablet Routines Questionnaire) which assess the daily routines for taking medication and the proportion of medication an individual has missed in the previous week and previous month5
General factors and adherence-enhancing strategies
Pill counts These are time-consuming, labour intensive and there is great potential for inaccuracy. Adherence (%) can be calculated as (the number of pills taken / the number of pills prescribed) × 1007
Many different factors affect adherence and these vary between patients. Consideration of these factors during a clinical assess ment should be the first stage of any intervention to enhance adherence for a patient (see Table 2).
Electronic methods Electronic devices have been developed which can be attached to the tablet bottle. They record the time and date on every occasion that the bottle is opened
Patient and illness characteristics Sociodemographics: systematic reviews of the literature high light that age at illness onset, age at first hospitalization, sex, socioeconomic status, marital status and ethnicity are not consis tently associated with differences in medication adherence.16,17
Prescription (refill) monitoring The frequency of prescription dispensing for an individual can be monitored as a proxy measure of adherence
Adherence history: partially adherent patients are more likely to have a past history of non-adherence than fully adherent patients.5 Thus, early detection and management of nonadherence may be of particular importance.
Saliva, plasma and urine assay tests These are the most objective measure of adherence but do not exist for all psychiatric drugs and are expensive and more invasive. Furthermore, assay tests have limited value in assessing partial adherence and may overestimate adherence with drugs that have a long half-life3,6,8
Illness duration: in chronic diseases, adherence usually declines over time and is generally lower when the patient is asymptom atic, treatment is prophylactic and the consequences of stopping treatment are delayed.2,8 However, in patients with first-episode psychosis, non-adherence rates can be high (40%) and beliefs about need for medication appear particularly important.
Table 1
Psychopathology: hostility, suspiciousness, disorganization and ‘hypomanic denial’ have been shown to be significantly associ ated with non-adherence in psychotic illnesses. However, global measures of symptom severity in schizophrenia do not appear to be consistent predictors of adherence.14,16
Magnitude and impact of non-adherence Non-adherence rates are reported as 40–60% for antipsychotics, 18–56% for mood stabilizers and 30–97% (median 63%) for antidepressants.5,6,10 Whilst these estimates might seem unac ceptably high, similar (if slightly lower) rates are seen in all other medical specialities. Medication non-adherence explains much of the difference between drug efficacy under ideal clinical circumstances and drug effectiveness seen in routine clinical practice.
General strategies for adherence enhancement
Relapse: non-adherent patients with schizophrenia are 3.5 times more likely than adherent patients to relapse within two years.8
• Questions about adherence should be routinely included in clinical assessment and repeated during follow-up, as adherence declines over time • Be alert to the impact of cognitive deficits, alcohol and illicit substance use on adherence • Simplify dosing frequency and polypharmacy where possible, and consider potential benefits of alternative drug formulations • Consider potential detrimental effect of frequency and severity of side effects on adherence. Titrate to minimal optimal dose for good efficacy but few side effects • Ensure instructions about medication are adequately understood by the patient and enlist family support where appropriate
Rehospitalization: non-adherence accounts for 40% of re hospitalization costs within 2 years.11 Time to remission: repeated relapse, often due to non-adherence, is associated with an increased difficulty or delay in achieving remission.12 Clinical burden and cost: non-adherence has a predicted excess annual cost of £2500 per patient for inpatient services and over £5000 for total service use.13 Attempted suicide: in both schizophrenia and bipolar disorder, risk of suicide attempt has been found to be significantly higher in patients not adhering to their medication regimens.12,14
PSYCHIATRY 6:9
Table 2
358
© 2007 Elsevier Ltd. All rights reserved.
Specific community treatments
Insight: illness beliefs are important predictors of adherence, although patients with no insight into illness may still accept and derive benefit from medication. Thus, helping patients to accept a mental illness label is less important than enhancing awareness of the impact of medication on their daily functional ability.2,16,17
Other significant factors Clinician: a poor clinician–patient relationship will adversely affect adherence. Clinician factors associated with adherence include: • the extent to which the clinician values medication • job satisfaction, i.e. morale and enthusiasm • continuity of clinician.4,8,16,18
Cognitive deficits and use of prompts: the patient’s ability to understand the purpose of medication and follow his or her medication regimen may be limited by cognitive deficits. Hence, cognitive and memory-enhancing strategies (e.g. external prompts such as telephone reminders) may be used to enhance adherence.2,8
Coercion: within the framework of the Mental Health Act 1983, psychiatrists can enforce medication against a patient’s will. Non-adherence is associated with the use of such measures, par ticularly with regard to involuntary admission. Thereafter, some patients make a smooth transition to voluntary treatment having directly experienced the benefits, whereas others remain resent ful. It can therefore be hard to establish a therapeutic alliance. Recent changes in law have created new orders called ‘super vised community treatment’ (England and Wales) and ‘compul sory treatment orders’ (Scotland), which can require compulsory medication adherence in the community. How effective these will be remains to be fully evaluated.
Comorbidity: alcohol and substance misuse are strong predic tors of non-adherence. Sometimes patients quote their clinician’s advice not to mix medication with alcohol as a reason for nonadherence.8,14,16 Medication and side effects Dose strength: the relationship between dose strength and adherence remains unclear but is probably curvilinear, with very low doses being associated with poor efficacy and very high doses with excessive side effects.8,16 High doses are sometimes prescribed because of apparent lack of pharma cological effectiveness, when the real cause is in fact partial adherence.
Clinic setting and organization: improvements to décor and ambience, offering refreshments and maintaining a realistic appointment schedule are all claimed to be associated with improved adherence.1,8 Access: lack of access to care and practical barriers (e.g. finan cial) have been highlighted, particularly for homeless persons, who are also often viewed as non-adherent. Hence clinicians should assist eligible patients in claiming for free prescriptions.7
Dosing frequency: it has been suggested that once-daily dosing is better than three times per day but that once-daily dosing is more frequently associated with partial adherence.4
Financial incentives (money, cash or vouchers redeemable for other goods) have been found to enhance adherence in substance abuse and non-psychiatric disorders, and are more cost-effective than other strategies. However, there are ethical concerns regarding the use of financial incentives for patients with major psychiatric disorders.
Drug formulation and route of administration: depot injections are associated with higher adherence rates than oral formula tions, drugs in liquid form higher rates than tablets, and sublin gual higher rates than normal tablets. Depot antipsychotics may be associated with lower relapse rates than antipsychotic tablets and afford early detection of non-adherence, thereby enabling early intervention. However, merely switching patients to depot without considering their reasons for oral non-adherence is unwise.4,8
Family support is associated with better adherence, although stressful social interactions and shared superstitions or false beliefs about medications may negate this effect. Generally, it is advoca ted that family support be enlisted. Alternatively, formal super vised medication administration – by members of the community mental health team (CMHT), for example – can be arranged.7
Drug packaging: calendar packs for tablets have been associ ated with better adherence than normal packaging (e.g. tablets in strips of ten).4 With a complex drug regimen, appropriate packaging may enable the patient to take the correct dose at the correct time. Alternatively, a ‘dosette box’ may be used.
Community-based interventions: components of community programmes have included provision of a strong and support ive social network and stable housing, other supportive services and close monitoring of clinical status. These can be effective in improving adherence in patients with psychosis.6,19
Polypharmacy: minimizing non-essential polypharmacy or com bination therapy (e.g. unnecessary anticholinergics or hypnotics) can improve intentional and unintentional non-adherence.
Psychological factors and adherence-enhancing interventions Factors underpinning non-adherent behaviour include the patient’s beliefs and attitudes concerning the illness and medication. Various psychological interventions specifically target these to enhance adherence (see Table 3).
Side effects: medication side effects are commonly cited by patients as their main reason for non-adherence, perhaps because clinicians consistently underestimate both the frequency and severity of side effects. As side effects do not appear to be a consistent predictor of non-adherence, it has been suggested that it is the fear rather than the actual experience of side effects that increases the risk of non-adherence.5,16
PSYCHIATRY 6:9
The health belief model The health belief model outlines four main belief categories: ben efits; costs; susceptibility; and secondary benefits of medication 359
© 2007 Elsevier Ltd. All rights reserved.
Specific community treatments
Family and group: family-directed interventions assume that the family unit influences the course of a patient’s illness. Such programmes usually include psychoeducation and are, in them selves, largely ineffective in enhancing adherence for schizophre nia, but strategies which involve the family are deemed more effective in bipolar disorder.14 Merely imparting knowledge that is not specific and personal to the individual does not necessar ily lead to a change in behaviour. Group interventions are based on the importance of peer support and shared identification and are usually based on psychoeducational principles. Again, these have largely been shown to be ineffective unless they also focus on behavioural change.6
Psychological interventions for adherence enhancement • Aim to understand the patient’s personal construct of the meaning of medication and illness • Consider the patient’s individual cost–benefit analysis of medication in the context of their personal goals and priorities • Be willing to acknowledge the potential positive value of symptoms and negative consequences of taking medication • Remember that simple psychoeducational programmes generally do not improve adherence • Consider using behavioural modification strategies (e.g. reminders)
Individual psychological interventions Cognitive–behavioural therapy (CBT) comprises two main com ponents and has been seen to be effective for adherence enhance ment in bipolar disorder.14 The cognitive component effectively targets the patient’s attitudes and beliefs towards medication and assumes that adherence is determined by the personal construc tion of the meaning of medication and illness. The behavioural component is based on the assumption that behaviour is learnt and can be modified, and so patients are provided with instruc tions and strategies (e.g. reminders, self-monitoring tools, cues and reinforcements); these can also improve adherence.6
Table 3
and adherence. It focuses on the patient’s decision-making pro cess, emphasizing the subjective cost–benefit analysis of any treatment, including probabilities of advantages and likely bur dens, in the context of the patient’s personal goals and priori ties. The underlying premise is that adherence enhancement is possible if the patient’s perceptions alter (see Table 4). However, this model may have limited predictive value for non-adherent behaviour. Application of the model to adherence-enhancing strategies can be compromised by a patient’s impaired cognitive function, limiting their capacity in risk assessment and planning. Moreover, a patient’s attitude to medication may be completely different from their actual medication-taking behaviour.2,6,8
Motivational interviewing enables the patient to express per sonal reasons for and against improving their adherence behavi our. Importantly, the clinician is required to acknowledge the potential positive value of symptoms and also the negative consequences of taking medication before the patient may even tually decide that symptom control by medication is preferable.2
Psychoeducational interventions Individual: many clinicians believe that patient education regarding illness, relapse prevention and medication will directly improve adherence. However, as most psychoeducational pro grammes focus primarily on imparting knowledge rather than on attitudinal and behavioural change, these have proved largely ineffective in enhancing adherence, although they remain more popular for bipolar disorder. Furthermore, increasing knowledge about illness and medication may be disturbing to the patient, as it may be a direct challenge to their only partial insight, although the patient’s fundamental right to information should not be compromised.6,8,14,17,18
Compliance therapy is a brief adherence-enhancing intervention based on motivational interviewing and cognitive approaches. Patients express their beliefs and concerns about medication and focus on personal goals as well as the more indirect benefits of medication, including improved personal relationships. Ambi valence towards medication is explored, consequences of medi cation cessation are discussed, analogies with chronic physical illness are made, and the pros and cons of medication are con sidered. The aim is to highlight the role of medication in staying well to fulfil needs and achieve lifestyle and personal goals. Com pliance therapy is advantageous in terms of insight, attitudes to treatment, improved social functioning and adherence, immedi ately post-therapy as well as at 18-month follow-up.3 However, in a more recent multicentre, European randomized controlled trial (RCT) comparing compliance therapy with a health edu cation intervention, participants in both groups improved. The study failed to show a benefit for compliance therapy over the control intervention, possibly because the control intervention also improved adherence.20
The health belief model Four main belief categories: • benefits • costs • susceptibility • secondary benefits of medication and adherence Adherence enhancement is improved if patients believe: • they are more likely to experience benefits from adherence • there will be few personal costs resulting from adherence • they will be at high risk of becoming seriously unwell if they do not adhere
Research limitations • Many adherence studies are subject to selection bias of re search participants and are also limited in the generalizability of their findings. Non-adherent patients are, by definition, more difficult to recruit to such studies, for the same reasons that they are non-adherent in the first place, such as homelessness.
Table 4
PSYCHIATRY 6:9
360
© 2007 Elsevier Ltd. All rights reserved.
Specific community treatments
• Retrospective study designs are commonplace, but accurate calculation of adherence rates and true identification of risk factors for non-adherence warrants prospective studies. • Most studies have concentrated on adherence only in psych oses or mood disorders. • There are numerous methods for measuring medication ad herence, with differing criterion thresholds used for defining non-adherence; there is thus little agreement between methods.15 Many methods do not truly measure if the patient actually in gested the medication and no single method is ideal. • Studies on adherence enhancement often involve complex in terventions in which the specific active component is not identi fied. They can be labour intensive and so may not readily be applied in everyday clinical practice. Studies often do not have a sufficient duration of follow-up to measure enduring effective ness, and measure adherence rates only, not clinical outcomes such as reduction in psychopathology.6,18
4 Demyttenaere K. Compliance during treatment with antidepressants. J Affect Disord 1997; 43: 27–39. 5 Scott J, Pope M. Nonadherence with mood stabilizers: prevalence and predictors. J Clin Psychiatry 2002; 63: 384–90. 6 Zygmunt A, Olfson M, Boyer CA, Mechanic D. Interventions to improve medication adherence in schizophrenia. Am J Psychiatry 2002; 159: 1653–64. 7 Azrin NH, Teichner G. Evaluation of an instructional program for improving medication compliance for chronically mentally ill patients. Behav Res Ther 1998; 36: 849–61. 8 Fenton WS, Blyler CR, Heinssen RK. Determinants of medication compliance in schizophrenia: empirical and clinical findings. Schizophr Bull 1997; 23: 637–51. 9 Gilmer TP, Dolder CR, Lacro JP, et al. Adherence to treatment with antipsychotic medication and health care costs among Medicaid beneficiaries with schizophrenia. Am J Psychiatry 2004; 161: 692–99. 10 Pampallona S, Bollini P, Tibaldi G, Kupelnick B, Munnizza C. Patient adherence in the treatment of depression. Br J Psychiatry 2002; 180: 104–09. 11 Weiden PJ, Olfson M. Cost of relapse in schizophrenia. Schizophr Bull 1995; 21: 419–29. 12 Leucht S, Heres S. Epidemiology, clinical consequences, and psychosocial treatment of nonadherence in schizophrenia. J Clin Psychiatry 2006; 67(suppl 5): 3–8. 13 Knapp M, King D, Pugner K, Lapuerta P. Non-adherence to antipsychotic medication regimens: associations with resource use and costs. Br J Psychiatry 2004; 184: 509–16. 14 Sajatovic M, Davies M, Hrouda DR. Enhancement of treatment adherence among patients with bipolar disorder. Psychiatr Serv 2004; 55: 264–69. 15 Dolder CR, Lacro JP, Leckband S, Jeste DV. Interventions to improve antipsychotic medication adherence: review of the recent literature. J Clin Psychopharmacol 2003; 23: 389–99. 16 Lacro JP, Dunn LB, Dolder CR, Leckband SG, Jeste DV. Prevalence of and risk factors for medication nonadherence in patients with schizophrenia: a comprehensive review of recent literature. J Clin Psychiatry 2002; 63: 892–907. 17 Colom F, Vieta E, Tacchi MJ, Sanchez-Moreno J, Scott J. Identifying and improving non-adherence in bipolar disorders. Bipolar Disord 2005; 7(suppl 5): 24–31. 18 Haynes RB, McDonald H, Garg AX, Montague P. Interventions for helping patients to follow prescriptions for medications (Cochrane Review). The Cochrane Library, Issue 3. Oxford: Update Software, 2003. 19 Nosé M, Barbui C, Gray R, Tansella M. Clinical interventions for treatment non-adherence in psychosis: meta-analysis. Br J Psychiatry 2003; 183: 197–206. 20 Gray R, Leese M, Bindman J, et al. Adherence therapy for people with schizophrenia: European multicentre randomised controlled trial. Br J Psychiatry 2006; 189: 508–14.
Conclusion Research which has attempted to identify general factors that predict variance in adherence has generated mostly contradic tory findings. However, psychological research attempting to understand why patients are not adherent has found attitudes and behaviours to be better predictors of non-adherence.4,8 Simple psychoeducational interventions are currently less successful in enhancing adherence than might be predicted and cognitive–behavioural techniques are more likely to succeed. However, dedicated time and resources for such interventions remain rare, despite high rates of non-adherence being common place. Such psychological therapies should not stand alone and employing a combination of strategies may be more favourable.15 Adherence-enhancing techniques should be routine for all clini cians, supported by the multidisciplinary team. Adherence studies underline the need to strive to understand our patients from their personal individual perspectives. This means giving them the opportunity to have their voices adequately heard. In so doing, it is hoped that the adverse clinical and eco nomic impact of non-adherence may be further reduced. ◆
References 1 Corrigan PW, Liberman RP, Engel JD. From non-compliance to collaboration in the treatment of schizophrenia. Hosp Community Psychiatry 1990; 41: 1203–11. 2 Hughes I, Hill B, Budd R. Compliance with antipsychotic medication: from theory to practice. J Ment Health 1997; 6: 473–89. 3 Kemp R, Kirov B, Hayward P, David A. Randomised control trial of compliance therapy. Br J Psychiatry 1998; 172: 413–19.
PSYCHIATRY 6:9
361
© 2007 Elsevier Ltd. All rights reserved.