Intended for healthcare professionals


Controversy over generic substitution

BMJ 2010; 340 doi: (Published 01 June 2010) Cite this as: BMJ 2010;340:c2548
  1. R E Ferner, consultant clinical pharmacologist12,
  2. Warren Lenney, consultant respiratory paediatrician3,
  3. John F Marriott, professor of clinical pharmacy4
  1. 1West Midlands Centre for Adverse Drug Reactions, City Hospital, Birmingham B18 7QH
  2. 2School of Clinical and Experimental Medicine, University of Birmingham, Birmingham B15 2TT
  3. 3Academic Department of Child Health, University Hospital of North Staffordshire and Keele University, Stoke-on-Trent, Staffordshire ST4 6QG
  4. 4School of Life and Health Sciences, Aston University, Birmingham B4 7ET
  1. Correspondence to: R E Ferner r.e.ferner{at}
  • Accepted 13 April 2010

Substitution of branded medicine with a generic equivalent is already common. Robin Ferner, Warren Lenney, and John Marriott argue that concerns about UK plans to let pharmacists make the decision are unwarranted

Consultation on the Department of Health’s decision to introduce generic substitution into English primary care completed at the end of March.1 The proposal, which is part of the 2009 pharmaceutical price regulation scheme for the whole of the United Kingdom, would allow pharmacists and other dispensers to fulfil a prescription for branded medicines by dispensing an equivalent generic medicine.2 Doctors will be able to over-ride substitution, and the department proposes producing a list of those drugs that can be substituted.

Generic substitution is already the norm in NHS hospitals, and 83% of NHS general practice prescribing in 2008 was for generic medicines. It is also used in Australia, the United States, and many countries in the European Union.3 4 Nevertheless, the UK proposal has been vociferously opposed by patient groups, doctors, and drug companies, sometimes in concert.5 Here we discuss some of the issues that make generic substitution in the community contentious.

Differences in content

Many generic drugs contain exactly the same active ingredient as the originator product. However, sometimes it is in a different chemical form, such as another salt. Erythromycin, for example, is available as ethyl succinate, lactobionate, stearate, and erythromycin base. The actions of different forms may not be identical, and they may have different solubilities and absorption characteristics. Dosages expressed as the weight of a salt rather than the weight of the active moiety (chemical entity) will also differ—for example, 200 mg ferrous sulphate contains as much elemental iron (65 mg) as 557 mg of ferrous gluconate. However, if the concentration at the site of action and rate of absorption of branded and generic products are the same, their clinical effect will also be the same.

Most medicines also contain a wide range of excipients—stabilising, bulking, flavouring, colouring, and sweetening agents. Differences between products can provoke allergy or intolerance. Excipients occasionally influence the acceptability of medicines; a recent reformulation of the Eltroxin brand of levothyroxine to contain sodium citrate was blamed for an outbreak of rashes and sore eyes in New Zealand.6

Licensing of generic products

Generic substitution is governed by national law, but in the European Union, marketing authorisation for a generic product is governed by European law. The manufacturer has only to show that the generic has pharmacokinetic properties similar to the originator, not to conduct clinical trials. Pharmacokinetic equivalence is taken to assure clinical equivalence.

A generic product is deemed equivalent to its originator if the 90% confidence intervals for the ratios of mean peak plasma concentration (Cmax) and extent of exposure to the drug (area under the concentration-time curve) lie between 0.8 and 1.25 when measured in at least 12 healthy volunteers (figure). This usually means in practice that pharmacokinetic values differ by less than 10% between generic and originator. Studies of generics in the US indicate that the differences for peak plasma concentration and area under the curve are generally 3-4%.7 The Danish Medicines Agency has narrowed the limits for bioequivalence to 0.90-1.11 for drugs with a narrow therapeutic range.8


Pharmacokinetics of hypothetical digoxin formulations, with the ranges of possible values for bioequivalence in EU (90% CI 0.8 to 1.25) and drugs with narrow therapeutic range in Denmark (0.9 to 1.11)

The EU limits for pharmacokinetic variation seem to be sufficient to ensure that the clinical efficacies of cardiovascular drugs are similar to those of the innovator product, but the situation is less clear for psychoactive drugs.9 10

Variability in therapeutics

Variation in doses occurs even in branded medicines. For example, by the British Pharmacopoeial standard, the quantity of insulin aspart (a drug that requires careful dosage adjustment) need only be between 90% and 104% of the stated amount. Three factors mitigate the effects of variation between products. Firstly, variation in bioavailability is usually small compared with the therapeutic variability within and between subjects. The standard dose of amoxicillin is recommended for any patient over 10 years old, even though he or she may weigh anything from 20 kg to 200 kg. Variation within individuals may be more important for long term treatments, but there is a tendency to underestimate both the variability in branded medicines and the factors that alter their effects from day to day, such as feeding or fasting, the presence of fever, and interactions with other medicines, food, and drink. For example, taking tetracycline with food halves its bioavailability,11 and taking erythromycin with grapefruit juice rather than water increases the maximum concentration and area under the curve by 50%.12

Secondly, the effects of a drug are related to the logarithm of the dose, so that small dose changes have little biological consequence. Thirdly, for most drugs, the margin between the lowest effective and the lowest toxic dose is wide and precise dosing is unnecessary. However, medicines with a narrow safety margin, such as anti-rejection drugs, previously encountered mainly in hospital, are now commonly dispensed in the community.

Financial benefits

Money is certainly one reason why generic prescribing could seem attractive to the Department of Health. Savings can be made—for example, Cipramil (£14.91/month) costs 11 times more than generic citalopram (£1.26/month). In the US the average generic price fell to 20% of the branded price when there were many generic manufacturers, although the fall was less if there was only one.13 The Department of Health expects that generic substitution in primary care will save £45m a year, which represents less than 0.5% of the annual NHS medicines budget.

Other benefits

Because medicine is international and many doctors work in countries where they did not train, it is helpful to use the recommended international non-proprietary names (rINN), which the British approved names (BAN) follow closely. Non-proprietary names are also designed to be informative. This reduces the risk of a patient being prescribed two different -statins or being given a -cillin to which he or she is allergic. The smaller number of names also makes confusion between drugs less likely.

In the US, generic substitution allied with free prescriptions increased adherence in some circumstances, implying that patients can be persuaded of the virtues of thrift if the benefits accrue to them directly.14 The Australian scheme, which allows patients to have branded medicines if they pay a brand supplement, also works well.3

Generic substitution could also allow pharmacy stockholding to be reduced. For example, there are five formulations of glyceryl trinitrate available (four branded and one generic). If a pharmacy stocked only the generic version of all drugs it would save money and make it easier to provide a rapid and uninterrupted service to patients.

Potential harms

The licensing regulations should ensure that generic products are similar to the originator product, and there are strong financial arguments in favour of their use. Why then do some oppose a relatively straightforward change in NHS service provision that would harmonise NHS community and hospital prescribing?

Some patients may find that originator and generic medicines differ in acceptability or efficacy. Children, especially, may dislike a formulation with an unfamiliar or unpleasant taste. Adherence in all age groups is already poor in chronic disorders: about half of adults prescribed an antihypertensive drug stop taking it within one year.15 Retail pharmacies are run for profit, and consequently seek out the lowest cost products, so suppliers often change. One consequence is that the physical form of a medicine can differ markedly from one month to the next, increasing patients’ confusion.

Generic manufacturers have become more numerous and more geographically dispersed. This has led to problems, such as the contamination of heparin with over-sulphated chondroitin sulfate.16 Vigilant quality control is essential to ensure confidence but will become more difficult as generic medicines are increasingly manufactured overseas. Regulators need to ensure that tests of bioequivalence are robust, perhaps by adopting more stringent conditions, as the Danish regulator has done; by requesting studies in patients as well as volunteers; and by continued supervision of products to ensure conformity.

The health of a small group of patients depends on drugs with a narrow therapeutic range such as anti-epileptic and anti-rejection drugs. Epilepsy Action argues that anti-epileptic drugs should be excluded from any generic substitution proposals. Since there will always be cases where seizure control is lost after a change from a branded to a generic preparation, whether or not the loss of control is caused by the switch, it may be pragmatic to concede this. Modified release preparations can have very different release properties, and the British National Formulary recommends that prescribers maintain brand consistency for several such medicines (table 1).

Examples of BNF recommendations to avoid brand substitution in drugs for which variations in bioavailability may alter clinical outcome

View this table:


Generic prescribing can help make the most efficient use of NHS resources. It is sometimes inappropriate, but the reality is that differences in drug dosage between one preparation and another are unlikely to be clinically important for most drugs and most patients, and doctors should explain this to patients. Tighter regulation would help to boost confidence in the process. We should applaud the current high rate of generic prescribing in the UK, and support the DH proposal to introduce an inclusive scheme, with suitable safeguards, as a way of consolidating generic substitution and improving rational and cost-effective prescribing.

Key terms

  • Generic prescribing—Prescribing by approved (non-proprietary) name

  • Generic substitution—Dispensing of an equivalent generic product containing the same drug(s) when a branded medicine is prescribed

  • Therapeutic substitution—Dispensing a different drug believed to have similar therapeutic actions to one that has been prescribed (eg, simvastatin in place of atorvastatin)

  • Bioequivalence—Where two medicines produce the same plasma concentrations of active ingredient, on which basis they are taken to have similar clinical effects

  • Clinical equivalence—Where two medicines produce the same clinical effect


Cite this as: BMJ 2010;340:c2548


  • Contributors and sources: REF has a longstanding interest in the safe, effective, and cost effective use of medicines. WL was a member of the joint standing committee of medicines (Royal College of Paediatrics and Child Health (RCPCH) and Neonatal and Paediatrics Pharmacists Group) which produced the first Medicines for Children Formulary in 1999, represents RCPCH on the strategy committee of the British National Formulary, and chairs the national paediatric formulary committee, which is responsible for the production of the British National Formulary for Children. JFM is a formulations consultant for the Medicines for Children Research Network and leads a research team devoted to the study of medicines management. He also has experience of medicines procurement in primary and secondary care. REF drafted the article after discussion with JFM and WL, who commented on and contributed to the article at each stage. JFM generated the data used to draw the figure. REF will act as guarantor.

  • Competing interests: All authors have completed the unified competing interest form at (available on request from the corresponding author) and declare (1) no financial support for the submitted work from anyone other than their  employer; (2) WL has received research grants, lecture fees, or travel expenses for meetings from GlaxoSmithKline, Novartis, Astra Zeneca, and Abbott; (3) no spouses, partners, or children with relationships with commercial entities that might have an interest in the submitted work; and (4) REF chairs a healthcare-wide drug and therapeutics committee that encourages generic prescribing.

  • Provenance and peer review: Commissioned; externally peer reviewed.


View Abstract