Article Text

Download PDFPDF

Can antibiotic use be both just and sustainable… or only more or less so?
  1. Michael Millar
  1. Correspondence to Dr Michael Millar, Department of Infection, 3rd Floor, Pathology and Pharmacy Building, 80 Newark Street, Whitechapel, London E1 2ES, UK; m.r.millar{at}


Antibiotic resistance threatens the capacity to treat life-threatening infections. If it is accepted that it will be many years (if not decades) until the production of new antibiotics overcomes current concerns with antibiotic resistance then ways to conserve the effectiveness of current antibiotics will have to be found. For many bacterial agents of infection levels of antibiotic resistance are directly dependent on the quantity of antibiotic prescribed. Antibiotics are currently underutilised in many parts of the world. If a just distribution of access to antibiotics requires equal access for individuals with equal need irrespective of wealth then responding to this requirement of justice has the potential to shorten the effective life of currently available antibiotics. Increasing the range and numbers of individuals treated with antibiotics would seem to threaten sustainability and also potentially undermine the access of future generations to cost-effective treatments for bacterial infection. The control of antibiotic resistance requires that the determinants of infectious disease transmission are addressed, such as poor housing, education and nutrition as well as the provision of antibiotics. The apparent tension between intragenerational justice and sustainability diminishes when the account of distributive justice extends beyond access to antibiotics and includes plural entitlements. Controlling antibiotic resistance requires more than the redistribution or reduction (in the overall use) of antibiotics.

  • Social control of science/technology

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

For the purposes of this paper antibiotics are considered to be drugs that can used to treat human infections caused by bacteria. The use of antibiotics can have benefits across a wide range of human activities including health care and farming; however, concerns have been raised for over 40 years that the use of antibiotics is leading to increasing pollution of the world with antibiotic-resistant bacteria.1 Antibiotic resistance reduces the effectiveness of antibiotic treatments. Much of current medical practice relies on the availability of effective antibiotics. Antibiotics are used to treat infections that arise as part of everyday life and also to prevent infections during periods when patients are particularly vulnerable (eg, during major surgery). A large proportion of hospital patients receive treatment with antibiotics, so hospital effluents may contain high concentrations of antibiotics, antibiotic-resistant bacteria and resistance genes.2

Cockell3 suggests that small creatures have not received the priority (in human concerns) that is given to large animals for reasons that may not be rational. Antibiotic-resistant bacteria have only recently begun to receive consideration in either the environmental or bioethics4 5 literature. Dwyer6 suggests that an important ethical task is to construct institutions and modes of living that promote health in ways that recognise the claims of sustainability and justice. I would like to consider the ethical implications of the relationship between the use of antibiotics and its consequence—antibiotic resistance for intragenerational and intergenerational justice.


Antibiotic-resistant bacteria are rapidly selected in individuals exposed to antibiotics and may persist for extended periods.7 Antibiotic resistance becomes a problem when resistant strains persist in individuals, human populations and/or in the environment to the extent that the efficacy of antibiotic treatment is compromised. It is currently estimated that in the European Union antibiotic-resistant bacteria cause 25 000 deaths a year, and contribute a yearly cost in US dollars of $21–34 billion.8

Historically antibiotic functions were considered as renewable resources, in that it was believed that if we stopped using a particular antibiotic then microbes would revert (back) to susceptibility, because maintaining defences against antibiotics was considered costly to microbes. Antibiotics may be a renewable resource in that ultimately microbes will revert back to susceptibility, but the time scale required for this reversion can be so long that the functions that particular antibiotics serve will not be available again for decades (eg, resistance of Staphylococcus aureus to penicillin). The costs of discovering, developing, testing and marketing new antibiotics have increased to the extent that antibiotic research and development is no longer attractive for many pharmaceutical companies. Few new classes of antibiotic have been discovered over the past 20 years, leading to various proposed strategies to promote drug company investment in the development of new antibiotics.8 If current patterns of high levels of antibiotic utilisation persist (while there remains a paucity of new antibiotics) then the availability of effective antibiotics will decline. It is unlikely that the problem of antibiotic resistance will be solved by new antibiotics in the near future—‘Today's dearth in antibacterial research and development will take decades to reverse’,9 so sustainability of the functions of antibiotics requires that we sustain the efficacy of those antibiotics that are currently available at least for the next few decades.

Sustaining the effectiveness of current antibiotics

There are a number of ways in which the effectiveness of existing antibiotics could be sustained. These include reducing the use of antibiotics, and/or reducing the factors that determine the transmission and persistence of antibiotic-resistant variants. We could potentially reduce consumption by using market forces, for example by increasing the price of antibiotics.5 We could regulate the use of antibiotics so that market forces driving the unnecessary or inappropriate use of antibiotics are curtailed.10 We can try to reduce the need to prescribe antibiotics by addressing remediable determinants of infection, for example by encouraging breast feeding, improving nutrition and improving access to and uptake of vaccines.11 We can try to control the factors that determine the transmission and persistence of antibiotic resistance such as by ensuring safe water supplies.

There are no existing and equally cost-effective alternative treatments to antibiotics for the majority of serious bacterial infections so, for example, the WHO and Unicef have prioritised the prevention and treatment of pneumonia in children less than 5 years of age, partly for the reason that the effective treatment of pneumonia is highly cost-effective with the potential to generate considerable returns in terms of disability-adjusted life-years averted.11 Currently (globally) pneumonia is estimated to kill 1.8 million children under the age of 5 years every year mostly in developing countries. Only one in five children under 5 years with pneumonia receive antibiotics, again mostly in the less affluent developing countries.11 Increasing the price of antibiotics is likely to prejudice further the access of those most in need to antibiotics. Is it just to take actions that lead to an increase in the price of antibiotics? If we choose to try to sustain the effectiveness of current antibiotics then which choices are compatible with justice?

Intra and intergenerational distributive justice

For the purposes of this analysis, I take intragenerational distributive justice to require comparisons of individuals from the same generation (eg, 2-year-old children in Europe with 2-year-old children in Africa) alive at the same time. Intergenerational justice requires comparison of different generations at the same chronological age (eg, 35-year-old adults born in 1900 and 1970). The availability of effective treatments for infection (antibiotics) is important now and is very likely to be important in the future, so sustainability is important both for intergenerational and intragenerational justice.

Intragenerational justice

The use of an antibiotic in an individual impacts on others and generates what economists refer to as externalities. These can be positive or negative. Antibiotic resistance as a consequence of the use of antibiotics is a negative externality. Prevention of the secondary spread of infection (by treatment of infectious individuals) with antibiotics is a positive externality. The extent to which these externalities are found depends on the context. A newborn baby has few bacteria from which to select antibiotic-resistant forms so the negative externality of antibiotic resistance will be quantitatively less than follows exposure of an adult to an antibiotic. The treatment of peritonitis with antibiotics in an adult (perhaps following appendicitis) does not give rise to positive externalities resulting from the treatment of infection because the causes of infection are already widespread (and part of our commensal flora), but does give rise to the negative externality of antibiotic resistance. The treatment of gonorrhoea in an individual contributes to protecting future contacts from gonorrhoea so carries both a positive externality and the negative externality of selecting for antibiotic-resistant bacteria. So the context is an important consideration when considering externalities arising from particular antibiotic treatment decisions.

Examples of recent ethical analyses have focused on describing the scale of the problem with antibiotic resistance, and potential strategies for controlling antibiotic resistance. Selgelid4 ascribes the burgeoning problem with antibiotic resistance to the failure of the market to control externalities such as antibiotic resistance. He argues that antimicrobial agents are ‘public goods’ that warrant special treatment and advocates governmental intervention and funding to direct the distribution of antimicrobial drugs. Selgelid4 points out that (globally) antibiotics are over-consumed by the wealthy and under-consumed by the poor. He argues that health is a ‘special good’, and that equal access to effective health care (and this would include effective treatments for infection) can be justified on egalitarian grounds, because health is required for normal species functioning12 or (it can also be argued) as an essential capability.13 Selgelid4 also argues that the control of the transmission of infectious disease by effective treatment of individuals (irrespective of wealth) can be justified for consequentialist reasons, which include control of the secondary spread of infections citing tuberculosis as an example. I agree with much of this analysis. Studies of the epidemiology of bacterial infectious diseases have shown that these diseases do not respect geographical or temporal boundaries. In the great majority of cases contact of an uninfected individual with an individual with an untreated bacterial infection (eg, tuberculosis, methicillin-resistant S aureus infection, or typhoid) increases the risk of infection of the uninfected individual. Effective control of many bacterial infections requires treatment of those who are infected so there are clear pragmatic reasons for ensuring the widespread and impartial availability of effective treatments for transmissible infectious diseases.

However, I do not agree with Selgelid4 that the under-consumption of antibiotics generally contributes to antibiotic resistance. Under-consumption may contribute to the selection of resistance in special cases such as the selection of resistant forms of tuberculosis. For many bacterial agents of infection levels of antibiotic resistance are directly dependent on the quantity of antibiotic prescribed (see, for example, Arason et al).14 Increasing the use of antibiotics in developing countries for the treatment of childhood pneumonia11 may well increase levels of antibiotic resistance. Poverty is significantly associated with an increased risk of many bacterial infections,15–19 and with the spread of infection both in developing20 and developed21 22 countries. Equal access for individuals with equal need irrespective of wealth has the potential to shorten further the effective life of currently available antibiotics. The implications for sustainability and intergenerational justice will be considered in the next section.

Anomaly5 advocates global treaties that would force up the effective price of antimicrobial agents so that the price includes the externalities associated with the generation of antibiotic-resistant bacteria. ‘The main obligation is… to make antibiotics expensive enough to curb low value consumption and encourage the development of new treatments in order to protect innocent parties.’ Anomaly5 acknowledges that increased drug costs will have a disproportionate effect on the poor, but states that ‘nobody has the right—not even the poor—to inflict uncompensated harms on other people against their will’ (p 18). If we accept that there are good egalitarian and consequentialist reasons for ensuring equal access to antibiotics based on need then use of price to curtail usage would seem to threaten both the control of infection and justice. Intragenerational justice requires access irrespective of wealth, and effective treatment of transmissible infectious diseases is of benefit to us all.

I agree with Anomaly that there are parallels between environmental pollution with antibiotic-resistant bacteria and with carbon dioxide. Bacteria (and other microbes) are involved not only as determinants of human (and animal) health but also play essential roles in many natural processes. The actions of individual consumers (of antibiotics or carbon-based energy sources) have a small impact on global pollution, but over time the persistence and accumulation of pollutants have serious consequences. There may even be interactions between the use of antibiotics and carbon dioxide levels. Antibiotics regulate microbial activities on many levels so have the potential to perturb naturally occurring microbial populations, for example in soil ecosystems. Microbes ‘are ubiquitous, possess enormous metabolic and physiological versatility and are essential to virtually all biogeochemical cycling processes—microbial carbon and nitrogen are calculated to be, respectively, equivalent to and tenfold as great as the carbon and nitrogen stored in plants’.23 Antibiotic pollution may be damaging these microbial ecosystems. However, parallels between antibiotic resistance and other forms of environmental pollution do not justify an argument for antibiotic treaties akin to those developed to control carbon dioxide or protect the ozone layer.5 Sources of energy and aerosol propellants can be changed but the functions of antibiotics cannot be replaced at the present time. Limiting carbon dioxide production by individuals is unlikely to damage their health irretrievably, whereas limiting the use of antibiotics for serious bacterial infection almost certainly will have significant health consequences for those deprived of access to antibiotics and potentially for all of us through the uncontrolled spread of infection.

If we accept that there should be equal access to effective treatments for bacterial infection based on need, then we will need to define and agree conditions under which antibiotics should be used and those under which antibiotics should not be used. I take appropriate use of an antibiotic to be use that can be justified by evidence of benefit (or good reasons to believe that there will be a benefit) to an individual or others (who may or may not be identifiable).

If one country and its people constrain the use of antibiotics so that they are only used in patients for whom there will be a substantial benefit, and another country uses antibiotics for the most trivial reasons or in excessive amounts or for unnecessary durations or to support unsustainable forms of animal husbandry, then we might judge the second country less well than the first one.24 If we accept that all countries should share the benefits of antibiotics, then we would probably also accept that all countries should be prepared to accept constraints on overuse.

Most would agree that a life-threatening bacterial infection would provide a justification for the use of antibiotics. Lack of access to effective antibiotics contributes to reductions in life expectancy (see, for example, Nugent et al),25 so could we simply calculate a sustainable level of global antibiotic usage per capita and multiply by population to give a ‘fair’ share per country?6 There are data on the outpatient use of antibiotics in European countries26 and this can be correlated with life expectancy. Life expectancy is very similar across the majority of countries in western Europe, yet the use of antibiotics varies widely between European countries so that Italy and France use far more antibiotics per capita than Germany, the UK or Holland. This result suggests that there may be differences in the efficiency with which antibiotics are used to improve life expectancy across the nations of Europe, but are France and Italy also using an unjust share of effective treatments for bacterial infection?

We cannot judge whether the use of antibiotics in a country is unfair without information on the contribution that the use of antibiotics makes to health or the context and reasons for antibiotics being used. Countries that use more antibiotics per capita may carry a larger burden of bacterial infections. Constraining the use of antibiotics without taking account of need might place considerable burdens on those communities with the greatest need, stifle developments and potentially have serious adverse consequences through the spread of infection. We need to compare like with like cases to see if these countries are taking an unjust share of the limited resource that is antibiotics. We need to agree criteria for the use of antibiotics.

Couper10 makes six recommendations to support the control of antibiotic resistance, and these include improving the availability of information and guidelines on the use of antibiotics, improved surveillance of drug resistance, regulation and assurance of drug quality, improvements in the education of prescribers and public, control of (inappropriate) promotional activities (by drug companies) and the encouragement of research for new antibiotics. I would add that in addition there should be minimum criteria for the use of antibiotics. The recent Center for Global Development (2010) report ‘The race against drug resistance’ makes four key recommendations.25 The third recommendation is that drug regulation in developing countries should be strengthened. I would argue that a requirement for equal access based on equal need requires international agreement and regulation of antibiotic prescribing across both developed and developing countries.

Justice can be conceptualised in many different ways. Tremmel27 identifies three conceptions of justice used in considerations of intragenerational justice. These are ‘justice as impartiality’, ‘justice as equal treatment of equal cases and unequal treatment of unequal cases’ and ‘justice as reciprocity’. An individual patient and/or doctor may believe that an antibiotic prescription will be of benefit and argue that individuals should have choices when it comes to antibiotic prescribing. There is tension between patient choice and public good when it comes to antibiotics. If we consider an impartial perspective then most would agree that effective antibiotics should be available to be used when there is clear evidence that the use of antibiotics would substantially improve health and/or control the spread of disease. On the other hand we would probably also agree that there will be circumstances in which the risks to others associated with the dissemination of antibiotic resistance outweigh potential benefits to individuals.

Minimum criteria for the use of antibiotics could be agreed so, for example, one criterion could be that antibiotics are prescribed by a doctor or in compliance with accepted guidelines, another that there are minimum agreed degrees of likelihood of benefit and scales of benefit required, and another that explicit reasons are given for prescribing. A substantial element within any agreed criteria for antibiotic use will be decisions about the degrees of benefit that justify the use of antibiotics set against the potential harms consequent on the selection of antibiotic-resistant strains. Criteria can be set that require account to be taken of empirical evidence, but antibiotic prescribing cannot be based solely on empirical evidence of effectiveness. For example, the use of antibiotics in patients with terminal illness is a complex and ethically difficult issue that cannot be decided on efficacy data alone. Minimum criteria for the use of antibiotics allow equals to continue to be treated equally and unequals unequally. There is mutual benefit in controlling pollution with antibiotic-resistant bacteria, so this type of agreement would also be compatible with a conception of justice as reciprocity.

Intergenerational justice

Compliance with accepted criteria for the use of antibiotics could be seen as just in an intragenerational context, but may still be unsustainable for succeeding generations. Quoting Dwyer6—‘the good of a relatively long and healthy life expectancy may be connected to harm done to the environment and, indirectly, to other people, nations, or generations’. If current patterns of antibiotic use cannot be sustained for future generations then can we describe those current patterns as just.

Present generations accrue the benefits of antibiotics and future generations carry the adverse risks associated with environmental pollution with antibiotic-resistant bacteria. For the purposes of this analysis I have accepted that we have obligations to future generations for reasons of empathy as much as any other.27 At the present time there is a paucity of new antibiotics and a burgeoning problem with resistance to existing antibiotics, so potentially from this point onwards each subsequent generation may be worse off than the previous one with respect to the treatment of bacterial infections. If we constrain antibiotic use to the extent that antibiotic effectiveness is sustained, then treatment of those alive today will be compromised, if we do not then in the absence of new antibiotics future generations may be blighted by untreatable infections.

If we take an impartial perspective and do not know to which generation we might belong then we would probably not want the benefits of antibiotics to be squandered. Modern medicine is using antibiotics at a level that threatens their effective use by future generations. However, there is an irreducible level of the use of antibiotics below which patient outcomes will suffer, and there are good reasons (both egalitarian and consequentialist) for not trading off morbidity and mortality of present generations for the sake of conserving antibiotic functions for future generations. Our responsibilities to ensure that future generations have access to effective treatments for infection requires that we take steps to conserve the functions of antibiotics while not prejudicing outcomes for patients at present. Allowing antibiotic prescribing without the fulfilment of minimum criteria is an abrogation of our responsibilities to future generations.

We do not know when new treatments (as effective as antibiotics) will be available to treat bacterial infections, and it is in large part because of this uncertainty that we have reason (at this present time) to be cautious in the use of antibiotics. On the other hand, progress continues—justice as intergenerational ‘equality’ is not an option because each generation starts from a different place. Each generation inherits innovations and inventions.27 Succeeding generations may have less access to fossil fuel sources of energy than we have, but will probably be able to utilise a wider range of energy sources as a consequence of technological progress. We do not know if this inequality will work in favour of succeeding generations with respect to the development of new antibiotics. Tremmel argues that ‘Inter-generational justice has been achieved if the opportunities of the average member of the next generation to fulfil his needs are better than those of the average member of the preceding generation’.27 If we are to fulfil this requirement of intergenerational justice then we will need to invest in research aimed at developing new treatments and preventive strategies for bacterial diseases. The current market model of drug development and marketing encourages high levels of use of new antibiotics while within patent protection. This model may need adjustment so that pharmaceutical companies are encouraged to see antibiotic development as a long-term investment and rewards adjusted accordingly.8

A second requirement if we are to try to move towards conserving the function of existing antibiotics is that we try to ameliorate the factors that determine the spread of bacterial infections (and resistant forms) such as poverty.15–19 Faden and Powers28 reject the idea that justice with respect to the distribution of health opportunities can be separated from other dimensions of wellbeing (p 3). They also argue ‘that empirical judgements of how various inequalities affect one another in concrete circumstances are ineliminable moral data’ (p 5). Providing antibiotics to children under the age of 5 years with pneumonia,11 without addressing other insufficiencies such as overcrowding, poor nutrition, education and inadequate sanitation, will only expedite the establishment and spread of antibiotic-resistant bacteria.20–22 The pressing problem with antibiotic resistance adds weight to those who argue for basic entitlements across a range of dimensions of wellbeing.29 13 28 The theory of justice based on capabilities proposed by Nussbaum13 bases justice on a requirement for core entitlements (capabilities). These entitlements are plural, mutually advantageous and outcome orientated. These entitlements require sufficiency across the range of chosen capabilities—sufficient for ‘a life worthy of human dignity’. ‘It then seeks political procedures (a constitution, various allocations of powers, a certain type of economic system) that will achieve that result as nearly as possible’. Controlling antibiotic resistance requires more than redistribution or reduction in the overall use of antibiotics. The control of antibiotic resistance requires that we address the determinants of infectious disease transmission as well as the provision of antibiotics. It is conceivable that in the future low (perhaps even sustainable) levels of antibiotic usage will be seen as a marker of a healthy society—a society with few of the currently prevalent risk factors for infection.

Extending the requirements of intragenerational justice to include minimum entitlements across a range of dimensions would do much to curtail the spread of infection and antibiotic resistance. Intragenerational justice understood in this way is less threatening to the sustainability of antibiotic effectiveness and intergenerational justice than justice narrowly construed as a just distribution of antibiotics.



  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Other content recommended for you