Are immediate- and extended-release drugs interchangeable?


  • Dan Cohen,

    1. Department of Severe Mental Illness, Mental Health Organization North-Holland North, Heerhugowaard, the Netherlands
    2. Department of Epidemiology, UMCG, Groningen, the Netherlands
    Search for more papers by this author
  • Anton J. M. Loonen

    1. Department of Pharmacotherapy, University of Groningen, Groningen, the Netherlands
    2. Department of Severe Mental Illness, Mental Health Organization Westelijk Noord-Brabant, Halsteren, the Netherlands
    Search for more papers by this author

  • Please also see editorial comments by David Taylor (Acta Psychiatr Scand 2013;127:6–7), Stephen M. Stahl (Acta Psychiatr Scand 2013;127:4–5) and Wolfgang Fleischhacker (Acta Psychiatr Scand 2013;127:8).

With the continuation of the euro crisis and the economic recession, budget cuts in all sectors of society, health care not excluded, are the call of the day. However, of all the costs involved – such as in-patient and out-patient care, loss of income, social benefit, criminal and juridical costs, cost of time and money spent by the family – the costs of medication are estimated at 10% at the most for the total expenditure on schizophrenia [1]. Nevertheless, substitution of branded drugs by their generic counterpart is popular as it promises an almost immediate reduction in costs without major disadvantages. The question we will discuss is whether this option is feasible with antipsychotic drugs used in the treatment of schizophrenia, schizoaffective and bipolar disorder.

Are immediate- and extended-release drugs interchangeable?

The case of carbamazepine. Comparison of four-times–daily immediate-release carbamazepine with the twice-daily extended-release epoxide formulation showed bioequivalence with no change in the seizure frequency [2]. In a head-to-head double-blind crossover study of conventional- vs. controlled-release carbamazepine, which included 48 patients with epilepsy, the controlled-release formulation was associated with a lower monthly seizure frequency – 9.3 vs. 6.3 (P = 0.013) – and a lower number of patients with side-effects: 26 vs. 6 (P < 0.001) [3]. The extended-release formulation also showed lower absorption variability compared with immediate-release formulation, which in the context of the narrow therapeutic window, especially for CNS side-effects [4], resulted in a lower incidence – 20% vs. 49% respectively – of CNS side-effects [5] The connection with psychiatry is two-fold. First, based on the results of two double-blind, placebo-controlled trials [6, 7], the FDA approved carbamazepine extended-release for the treatment of acute manic or mixed episodes of bipolar I disorder. Second, the results indicated that extended-release formulations reduce the burden that drug-therapy places on the patients, a point that is possibly even more important in patients with severe mental illness (SMI), such as bipolar disorder or schizophrenia, as it is in the general population. In short, the outcome suggests that immediate- and extended-release forms are not interchangeable. Second, both greater tolerability and lower burden on the patient suggest that extended-release formulations are the preferred choice.

Non-adherence and dosing frequency in chronic somatic disease

In general, chronic health conditions require long-term medication for optimal management. Medication frequency is one of the factors that influence medication adherence. In a systematic review of treatment adherence in chronic diseases such as diabetes mellitus, hypertension and stable angina, a relationship between daily medication frequency and medication adherence was found: 13–26% more adherent days in patients receiving once-daily dosing compared with twice-daily dosing and 22–44% more adherent days in once-daily dosing compared with thrice-daily dosing [8].

Non-adherence in schizophrenia

Non-adherence in schizophrenia is a common phenomenon. In a meta-analysis of 24 antipsychotic medication studies, overall non-compliance was estimated to be 42% [9]. It has been argued that the methods used to estimate compliance may have underestimated non-compliance [10]. In the clinical antispychotics trials interventions effectiviness (CATIE) study, any cause discontinuation of prescribed antipsychotic drug in after 1 year varied between 50% and 75% [11].

Non-adherence in schizophrenia is not only frequent, but also dangerous as evidenced by a drug dispensing data study that compared patients with a drug holiday – defined as no refill at the pharmacy for 30 days or longer – with patients without a drug holiday. Corrected for age and gender, the relative risk for suicide attempts in the group with drug holiday was 4.2 compared with those without a drug holiday [12].

In a long-term follow-up study (mean: 11.4 years) of 870 young (<30 years at index admission) patients with schizophrenia, seven times as many non-adherent patients committed suicide (OR: 7.0; 95% CI: 2.5–28) [13]. In another study in 1906 patients with schizophrenia spectrum disorders, improved adherence was associated with better treatment outcomes, including reduced rehospitalisation and reduced incidence of suicide attempts [14].

Treatment adherence: attitude and substitution

No substitution study of measured (non-)adherence behaviour in patients with schizophrenia or bipolar disorder is available. A single-blind simulation study of the effect of pharmacists initiated substitution of brand medication by its generic form on patients willingness to take the drug comes nearest [15] The treating psychiatrists had not discussed substitution with their patients; the patients were therefore unprepared for substitution that was initiated by the pharmacist. Substitution affected negatively on the patient's willingness to take the medication: of the original 81.5–90.4% willingness, 27% remained willing after unannounced brand-by-generic substitution; 52% wanted to consult their psychiatrist and 10% suspected a pharmacist mistake. This outcome suggests a negative impact of pharmacist initiated substitution on treatment adherence in unprepared patients.

Substitution: patient complaints

The European Medicines Agency (EMA) provides manufacturers with bioequivalence guidelines for the assessment of differences between branded and generic drugs [16]. The EMA guideline calls for studying characteristics in groups of healthy volunteers or, if the drug carries safety concerns that make this unethical, of patients: it demands the calculation of population-characterizing measures such as mean values and standard deviations of the peak plasma concentration (Cmax), and of the area under curve. For these parameters, the 90% confidence interval for the ratio of the test and reference products should be contained within the acceptance interval of 80.00–125.00%. In specific cases of products with a narrow therapeutic range or for highly variable drug products, the acceptance interval may be tightened (90.00–111.11%) or widened (69.84–143.19%), respectively. These parameters apply for a whole population, not for an individual. Evidently, not every individual patient has the group characteristics as is witnessed by patients who, within this frame of objective and measurable bioequivalence, still report differences in their subjective experiences of branded compared with generic drugs. The temptation to reject these claims as psychologically driven is nearly irresistible but not necessarily correct. Similar experiences are reported in the treatment of epilepsy with seizure frequency as an objective outcome [17]. Although also in this field, most reports on the loss of seizure control (or other adverse experiences) after switching generics for brand-name antiepileptic drugs are anecdotal and difference in efficacy cannot be concluded from meta-analysing randomized controlled trials, intensive monitoring is advised after switching in high-risk patients [18]. We suggest taking psychiatric patient's reports seriously and at face value for the following reasons. Psychological aspects of being properly treated (positive expectations, confidence in the healthcare providers) are significantly contributing to treatment results in many psychiatric conditions, and loss of trust may limit this response. Particularly, patients with a SMI can be considered to be high-risk patients, because treatment failure (e.g. due to non-adherence) may result in aggression and hostility. A discovery of relevant individual pharmacokinetic difference that explains the patients' subjective experiences might instead increase the credibility of the doctor and strengthen the therapeutic alliance between doctor and patient. Another reason to take such patient's report seriously is because it shows proper medical behaviour, as it implies consistency with the same scientific approach as shown to other problems, we are faced with daily practice. Indeed, it offers the opportunity to identify gaps in the current criteria and to improve them.

In short, when a patient experiences differences, it might signify that this patient does not fit into the well-documented group characteristics associated with this drug; this patient may simply display individual susceptibilities that differ from the general group characteristics. Perhaps, it is better not to check for difference between the averages, but to look for extravagancies. This should be the result of a careful postmarketing monitoring programme that compares the number of adverse experiences during monitored release.

Substitution: tolerability of immediate- and extended-release

Prospective head-to-head comparison of tolerability of immediate- vs. extended-release forms of the same drug, manufactured by the same company, is rare. In one such study, a double-blind crossover study of immediate and extended-release quetiapine in 58 healthy subjects, the objective pharmacokinetic data matched nicely the subjectively experienced drowsiness: increased plasma concentration during the first four hours in subjects taking immediate-release quetiapine were accompanied by increased drowsiness in the same period in the same population. In short: healthy subjects experienced lower intensity of self-reported sedation in quetiapine XR compared with quetiapine IR [19].


Non-adherence is a problem common to all chronic diseases. Substitution of generic for brand drugs in patients, with whom this topic has not been discussed, seems to affect treatment adherence negatively. This is the more worrying as non-adherence in schizophrenia is associated with increased suicide attempts and completed suicide.

The financial gain of substituting non-brand for branded drugs should not be ignored, but when considering the possibility of such a change, the disadvantages such as increased non-adherence with accompanying dangers of increased costs due to extra hospitalizations and/or suicide attempts should be factored in the decision-making process.

Finally, when considering immediate- vs. extended-release formulations, few comparative studies are available. However, studies that used carbamazepine or quetiapine immediate- and extended-release formulations suggest better tolerability with the extended-release formulations. As once-daily dosing is associated with higher adherence compared with twice-daily of three-times–daily dosing, both findings favour use of the extended-release forms.