Cost-Related Non-Adherence to Prescribed Medicines: What Are Physicians’ Moral Duties?

Abstract

As the price of pharmaceuticals and biologicals rises so does the number of patients who cannot afford them. In this article, we argue that physicians have a moral duty to help patients access affordable medicines. We offer three grounds to support our argument: (i) the aim of prescribing is to improve health and well-being which can only be realized with secure access to treatment; (ii) there is no morally significant difference between medicines being unavailable and medicines being unaffordable, so the steps physicians are willing to take in the first case should extend to the second; and (iii) as the primary stakeholder with a duty to put the individual patient’s interests first, the medical professional has a duty to address cost-barriers to patient care. In articulating this duty, we take account of important epistemic and control conditions that must be met for the attribution of this duty to be justified.

Keywords:

• professional ethics
• health care delivery
• moral theory
• pharmaceutical industry
• professional-patient relationship
• right to health care

INTRODUCTION

Medical Non-Adherence

The World Health Organization defines medical adherence as “the extent to which a person’s behavior—taking medication, following a diet, and/or executing lifestyle changes, corresponds with agreed recommendations from a health care provider” (World Health Organization 2003). When patients fail to adhere to medical recommendations, it not only affects them personally in terms of potentially worse health or lost opportunities for improved health, but is likely also to impact the health system via increased costs of care, hospitalization rates, and wastage of limited medical resources (Cutler et al. 2018; Van Alsten and Harris 2020; While 2020). In the United States, it is estimated that medical non-adherence costs hundreds of billions of dollars per year (Bosworth et al. 2011). Factors known to influence non-adherence include beliefs and perceptions regarding the necessity, adverse effects, and efficacy of treatment; the quality of the patient-physician relationship; and the complexity and cost of treatment (Mukhtar, Weinman, and Jackson 2014). Non-adherence can be deliberate and patient-motivated, or unintentional and driven by lack of patient capacity or resources (Molloy et al. 2014). According to WHO, improving treatment adherence could have a far greater impact on the health of the population than additional innovation (World Health Organization 2003).

Cost-related non-adherence (CRNA) to medicinal treatment is a particular challenge due to increasing medicine prices (Beasley 2023), increasing out-of-pocket spending on medicines (Australian Medical Association 2019), and a cost-of-living crisis imposing financial pressure on patients (World Economic Forum 2022). In Australia—a high-income country with a robust public health insurance scheme—5.6% of people in any given year either refrain from or delay taking medicines prescribed by their general practitioners for financial reasons (Australian Bureau of Statistics 2022). Certain groups are more significantly implicated. One in ten (9.5%) patients with poor or fair health, and up to 8.5% of young patients, are non-adherent with GP prescriptions for reasons of cost (Australian Bureau of Statistics 2022). In the United States patients pay comparatively more for medicines than anywhere else in the world, and unsurprisingly, CRNA is higher there than elsewhere. For example, in the US 25% of patients with chronic liver disease (Lago-Hernandez et al. 2021), 16% of patients with inflammatory bowel disease (Nguyen et al. 2021), 18% of non-elderly patients with diabetes (Taha et al. 2022), and 13% of patients with atherosclerotic cardiovascular disease, report CRNA (Khera et al. 2019).

The recent controversy surrounding the cost of diabetes treatment in the United States illustrates this challenge. The financial burden caused by high insulin prices (costing some individuals up to US$1,000 per month) led the government to implement a price cap of $35 per month as of 2023 for Medicare-eligible patients (Hirsch 2022). Prior to this intervention one study at the Yale Diabetes Center found that one in four people underused insulin for financial reasons, leading to poor glycemic control (Herkert et al. 2019).¹ An analysis of the US National Health Interview Survey Data found that CRNA to diabetes, cardiovascular, and hypertension medications was associated with a 22% higher risk of all-cause mortality (Van Alsten and Harris 2020). Patients struggling to achieve long-term insulin security described their situation as a “life and death” emergency (Willner, Whittemore, and Keene 2020). This is despite the fact that insulin was discovered over 100 years ago and should by now be widely accessible.

Physicians’ Duty to Promote Treatment Adherence

Despite the significant literature on non-adherence and its causes, and proposed strategies to ameliorate it, researchers and scholars are surprisingly silent about whether physicians have an explicit moral duty to address CRNA. In a recent article, Ghinea argues that physicians are duty-bound to help patients import more affordable medicines on the grounds that without such support, the most financially vulnerable patients will be further disadvantaged (Ghinea 2022a).

Other scholars have considered the related question of whether physicians have a duty to consider patients’ socioeconomic circumstances when making a treatment plan. For instance, Puschel and colleagues argue that the traditionally-conceived neutrality of the medical profession with regards to patients’ socioeconomic status is unjustifiable and reliant on an overly narrow conception of fairness (Puschel, Furlan, and Dekkers 2015). They propose that physicians be required to provide similar interventions to similar patients (regardless of socioeconomic circumstances), and by extension treat patients who differ meaningfully from each other differently - for instance, by giving special attention to those with more acute clinical needs. Furler and Palmer argue that physicians have a duty of “being-for” and “being-with” their patients; that is, physicians must have a reflexive awareness of the myriad ways in which they can respond to the needs of their patients (Furler and Palmer 2010). They bemoan the paucity of ethical guidance for physicians on how to respond to socioeconomic inequities, which, they say, “is surprising given the quotidian nature of this issue”.

In addition, Rogers has proposed that physicians must offer patients realistic management plans, and that those being treated are in the best position to evaluate whether or not the proposed interventions are realistic for them (Rogers 1999). This requires disclosing information that may help patients make this evaluation, including discussing the financial cost and potential outcomes of various treatment options. Concerned about demanding too much from physicians, Gruen and colleagues have sought to distinguish between physicians’ professional obligations and professional aspirations (Gruen, Pearson, and Brennan 2004). They propose that physicians have only a duty to address socioeconomic factors that have a direct impact on health and that they can directly modify (e.g. smoking), while other factors that influence health, such as education and income, are deemed to be “broad” socioeconomic influences that cannot feasibly be modified by individual physician intervention.

Other scholars, such as Hurst, believe that physicians should remain neutral regarding socioeconomic influences altogether, at least when it comes to prioritizing individuals for treatment (Hurst 2009). To do otherwise would, Hurst believes, amount to using medicine as an instrument to compensate for injustices that arise outside of medicine, and to apply different values within medicine than are applied in broader society. Pellegrino is similarly concerned with preserving a more narrowly-conceived telos for medicine, and believes that shifting the focus from patient-centric to socio-centric goals might lead to medicine becoming an instrument for social and political ends (Pellegrino 2001). However, a careful reading of Hurst’s and Pellegrino’s positions shows they are not antagonistic to the idea that physicians should help patients overcome cost-barriers to treatment. Rather, they appear primarily concerned about socio-political goals creeping into the practice of medicine in a way that could distract from a primary commitment to the needs of individual patients. In this regard it is worth noting that while many medical schools teach students about the social determinants of health, it is a low priority for various reasons including the belief that the issues raised are outside the realm of physicians’ responsibilities (Lewis et al. 2020).

AIM AND APPROACH

On what grounds, then, might it be argued that physicians have an obligation to identify and address CRNA to medicines? Our aim is to argue that addressing CRNA is not merely a widely held professional aspiration, as per Gruen et al.’s framing, but is rather a professional duty grounded in more general duties that physicians already accept. Furthermore, we will propose that in some circumstances this duty may require doctors to adopt potentially controversial strategies such as prescribing cheaper off-label alternatives in preference to on-label medicines, and recommending or facilitating importation of cheaper medicines from abroad.² In making this argument we recognize that there are certain epistemic and control conditions that are antecedent to this duty, and later in this paper we explore their implications for our conception of the relevant physician obligations.

ANALYSIS OF EXISTING AND WIDELY ACCEPTED MEDICAL DUTIES

Interpreting the Duty of Care: Duty to Offer Clinical Advice Alone versus Duty to Help Ensure Realization of Health Goals

Broadly speaking, the profession of medicine exists to improve health and wellbeing, and by extension each individual physician has a prima facie duty to do what they can to improve the health of their patients.³ However, the extent and limits of this duty remain open to debate. On a narrow account, physicians are obliged only to offer advice and interventions that they have the expertise to recommend. Insofar as medical training focuses on acquisition of medical skills and knowledge, this means physicians are primarily responsible for the deployment of these specific skills and knowledge when treating patients. Gruen et al. appear to allude to these specific skills and knowledge when justifying their distinction between professional duties and aspirations (Gruen, Pearson, and Brennan 2004). A key element of this distinction is the extent to which a responsibility qua physician differs from the responsibility of other citizens, groups, or institutions in society. For instance, according to Gruen et al., duties related to preserving equity and opportunity are broad public duties that extend beyond the proper scope of individual medical professional duties.

While there is merit in Gruen et al.’s attempt to circumscribe physician obligations, it is not plausible, we believe, to regard physicians as entirely absolved of responsibility for considering the impact of non-clinical factors on the health and wellbeing of their patients. Indeed this is acknowledged in various authoritative sources. For example, codes of ethics and conduct standardly recognize that physician responsibilities extend beyond purely clinical concerns, and include supporting and promoting access to care and helping patients achieve their health goals through various clinical and non-clinical means (American Medical Association,2023a 2023b; Medical Board of Australia 2023). Physicians are also legally responsible for taking reasonable steps to mitigate foreseeable risk of harm to their patients, where the focus is on the reasonableness of the steps rather than the types of steps taken (Ghinea 2023). Hence, where it can be foreseen that a patient is likely to come to harm because they cannot afford to buy essential medications, it seems to lie properly within their doctor’s scope of duties to take reasonable steps to identify and address such a risk.

In their accounts, Pellegrino and Hurst express concern that holding physicians responsible for socioeconomic determinants of health would be deleterious to medicine on the grounds that it would potentially compromise its telos (Hurst 2009; Pellegrino 2001). Yet they ignore the paradox that the very telos they seek to defend demands that physicians take account of these kinds of determinants. Clinical advice only has value insofar it contributes to improving health or preventing or slowing down further deterioration. The ability to adhere to prescribed treatment is a necessary condition for effective care. If CRNA starts to undermine the efficacy of clinical practice, then it surely falls within the telos of medicine and as such demands attention and response from doctors.⁴

In addition, we do not believe doctors are solely bound to uphold those duties distinct to their profession, as Gruen et al. seem to believe. Membership of a profession does not entirely absolve one of all broader and more commonly shared duties. For instance, while it could be argued that responsibility for addressing medicine affordability rests primarily with governments and insurers rather than physicians, it does not follow that physicians have no duty whatsoever to consider affordability when relevant to their patients’ capacity to access treatment. Furthermore, Gruen et al.’s approach fails to recognize that the causes of what they call “broad” socioeconomic influences are themselves complex and so demand a coordinated response. Hence, while we agree that physicians cannot be held responsible for solving the deeper systemic problems of inequity, we do believe they have a role to play in mitigating those dimensions of the problem over which they do have some degree of control—such as helping individual patients find solutions to cost-barriers to access.⁵

Be that as it may, even if one were to hold that physicians’ moral duties are limited to provision of clinical services and not to achievement of broader positive health outcomes for their patients, it is surely pertinent that many physicians feel the contrary. In their study examining perceived responsibilities and self-blame among diabetes patients and physicians, Beverly et al. found that interviewed physicians did self-attribute responsibility for patients not achieving their health goals (Beverly et al. 2012). Rogers similarly found that physicians she interviewed felt a responsibility to help patients navigate the “maze of going down the right tracks” inclusive of considering the temporal and financial aspects of care (Rogers 1999). Furthermore, in their study of humanistic values in medicine, Chou et al. found that many physicians had a relatively expansive view of the moral norms that apply to their practice, and that these included treating patients as one would want to be treated oneself, and ensuring that the needs of the entire person were addressed, not only those directly related to ameliorating disease (Chou, Kellom, and Shea 2014).

In summary, we believe that a responsibility for CRNA cannot be refuted by appeal to a narrow account of the duty of care whereby physicians are responsible solely for clinical advice with no consideration of the patient’s capacity to implement that advice in practice. Not only would such a narrow conception be inconsistent with existing medical codes of conduct and the law that frames professional responsibilities, it would also inadvertently support ineffective or suboptimal care, insofar as prescribing medicines patients cannot afford leads to them either forgoing, delaying, or rationing treatment in ways that are contrary to medical recommendation.

Interpreting the Duty to Act in the Patient’s Best Interests: Duty to Offer Most Effective Treatment versus Duty to Offer Affordable Treatment

We take for granted that physicians’ moral obligation to act in patients’ best interests requires them to advise patients of the most effective treatments available. Accordingly, a physician who knowingly withholds this information while offering an inferior medicine, or a medicine that has not been tested as rigorously as other available medicines, thereby breaches their duty.

Such a view seems to underlie Novartis and Bayer’s UK-based objection to the use of the much cheaper but off-label drug bevacizumab (Avastin) for aged-related macular degeneration (AMD), instead of the more expensive and on-label alternative (Haughey 2019). They argued such use was not appropriate on several grounds, including that bevacizumab was not licensed for that use and that prescribing it undermined patients’ right to have an approved drug. The UK General Medicine Council appeared to support this position by advising physicians not to prescribe off-label and unlicensed medicines when on-label alternatives were available (Dickson 2015). Presumably the GMC advised this because they believed off-label medications lack an equivalent evidence-base for benefits and harms and are therefore riskier. Nevertheless, in the case of bevacizumab for AMD, the UK High Court disagreed with this premise, ultimately supporting the off-label use on the basis there was no clinically meaningful difference in terms of efficacy or safety (Haughey 2019).

The key question for our purposes, however, is whether any difference in efficacy or safety would necessarily make prescribing a potentially less effective medicine for cost reasons ethically unacceptable. We don’t believe so, on the following grounds. The first step in our argument is to draw attention to the (non-cost-related) observation regarding unmet medical need, namely that physicians accept it is sometimes necessary and justifiable to incur certain risks, provided they are deemed proportionate to the expected benefits. For example, in some cases there may be no approved treatment for a specific condition, potentially warranting the use of off-label treatments to address this gap in spite of lack of evidence of efficacy (Ghinea et al. 2012; Largent, Miller, and Pearson 2009). A good example was the controversial uptake of the novel (at the time) medicine epticog alpha for the treatment of uncontrolled bleeding, whereas its licensed use was for the management of specific blood clotting disorders (Ghinea et al. 2012). Physicians deduced that its powerful coagulating effect offered a plausible basis for off-label use despite the lack of supporting evidence. Similarly gabapentin, originally approved as an anticonvulsant, became widely used off-label for the treatment of neuropathic pain due to etiological similarities between the two conditions and lack of treatment options for the condition at the time (Ghinea, Lipworth, and Kerridge 2015). Both medicines ended up being used almost exclusively off-label despite a lack of formal evidence of benefit. In addition, regulators also recognize that serious unmet healthcare needs may warrant releasing medicines onto the market earlier than would otherwise be the case. Such provisional approval schemes provide early access to medicines despite incomplete evidence of safety and effectiveness (Pace et al. 2017). The rapid rollout of COVID-19 vaccines during the pandemic is an exemplar of this approach (Doshi 2021). In cases such as these, unmet medical need provides the justification for accepting the risk of using medication for which there is less evidence than would standardly be required for full regulatory approval.

Extending the reasoning in this first step to the question of cost-related non-adherence, we suggest that from a patient’s perspective there is no practical difference between a situation in which they lack treatment options and one in which they cannot afford the available options; in both cases the patient has an unmet medical need. As a result, in situations where patients face cost barriers we believe it should similarly be ethically justifiable for physicians to prescribe cheaper alternatives, even where the prescribed medication may be less effective than more expensive alternatives, or where the safety of the cheaper preparation is less well established. It seems to us plausible that in at least some such cases, any risk of harm would be outweighed by the benefit to the patient of accessing some medication rather than none. An illustrative example is the use—across many low and middle income countries, and even in low-resourced areas of high income countries—of misoprostol as a cheap and accessible treatment for post-partum hemorrhage, despite this being an off-label application (Ghinea, Lipworth, et al. 2015). Misoprostol is prescribed in these cases because the gold-standard treatment oxytocin (which is also off-label) is either too expensive or inconvenient due to cold-chain storage requirements and the need for intravenous administration. Notably, using misoprostol in settings where oxytocin is not accessible is supported by the World Health Organization and the International Federation of Gynecology and Obstetrics (International Confederation of Midwives, and International Federation of Gynecology and Obstetrics 2014; Ghinea, Lipworth, et al. 2015).

Second, if physicians were strictly required only to prescribe the most effective available treatment, the prescribing and subsidization of any previous and slightly less effective medicines would immediately become obsolete on release of a more effective, and likely more expensive, product.⁶ But this does not occur; governments, insurers, and health services take account of the incremental cost-effectiveness of new medications before deciding to subsidize it. Indeed, to demand that physicians only prescribe the most efficacious treatment would be to bankrupt health systems—it would be impossible to financially support such a proposal. As a case in point, when sofosbusir (“Sovaldi”) was first released in the United States for the treatment of hepatitis C, it was estimated that funding it for all hepatitis C patients would cost the equivalent of the country’s entire pharmaceutical budget (Roy 2014).

On both pragmatic and consistency grounds, therefore, we suggest there is no requirement for physicians to offer only the most effective treatments to patients. Rather, the duty to act in the patient’s best interests demands that physicians offer patients the safest and most effective medicines accessible to them.⁷ And any accessibility assessment must in turn incorporate considerations of affordability.

Physicians’ Duty to Individual Patients

The third basis upon which we argue that physicians have a duty to help patients overcome cost-barriers to care and medical treatment is the fact that although many different parties are typically involved in the successful delivery of care, physicians represent the only healthcare profession with general prescribing rights that has a specific duty to serve individual patients’ best interests.⁸ Public hospitals, public insurers, and regulators owe duties to populations rather than individuals, and so must at times sacrifice individual interests to serve those of the broader community. Likewise, pharmaceutical companies, private insurers and other commercial care providers have an overriding fiduciary duty to their shareholders, meaning that they have a legal duty to serve their interests first and foremost, rather than the interests of patients.

As such, physicians can be understood to be de facto guardians of patients’ interests in a complex landscape driven by competing priorities and vested interests. In recognition of this, the Corporate Practice of Medicine Doctrine—which sets out allowable employment relationships between medical professionals and corporations to ensure that physicians’ care for their patients is not undermined—is still applied in most of the United States (Schaff and Krisza 2022). It is also in recognition of this that physicians are permitted to prescribe off-label in response to market failures that create a gap between clinically necessary treatment and commercially-driven drug development. This gap is so prevalent in relation to some specialties and conditions that off-label uses can constitute over half of all prescriptions (Ghinea, Kerridge, et al. 2017).

Based on the preceding arguments, we maintain that physicians are the primary candidates within healthcare systems to support individual patients’ access to affordable medicines. High rates of CRNA create a systemic pattern of ineffective prescribing that is to a significant extent avoidable (Bosworth et al. 2011). Those charged with prescribing medicines appropriately bear some of the burden of responsibility for addressing this problem.⁹

LIMITS TO THIS DUTY

We have argued for a specific duty on the part of physicians to identify and address medicinal CRNA for individual patients. However, no duty is absolute, and countervailing factors that may interfere with the ability of an individual to fulfill a duty must always be considered. It is widely accepted that for responsibility to be fairly attributed to an individual or entity, they must be able to meet both epistemic and control conditions (Levy 2018). Below we use a hypothetical scenario to illustrate and explore how these conditions may generate exceptions to the physician responsibility we propose.

Epistemic Grounds

Imagine that an oncologist specializing in the treatment of melanoma is not aware of the latest therapeutic developments in their field and so, when offering treatment, fails to disclose the most effective treatment according to the latest research. Since we would expect oncologists to know the best treatment options, along with their relative merits and disadvantages, we may consider the oncologist blameworthy for this lack of awareness. This example shows that we have epistemic expectations of professionals, and when these expectations are not met we can legitimately attribute blame.

In our preceding analysis we have proposed that physicians have a duty to help patients access alternative treatments that they can afford if confronted with cost-barriers to the licensed and/or clinically superior treatment. However, two assumptions warrant scrutiny: first, that the treating physician is aware (or ought to have made themselves aware) of a specific patient’s financial circumstances; and second, that the treating physician possesses knowledge about the costs, effectiveness and relative cost-effectiveness of the various treatment options. Knowledge of potentially cheaper alternative treatments is relevant in cases where it is known that a specific patient faces cost-barriers to care, so the first question is: are physicians required to be cognizant of their patient’s capacity to afford prescribed treatment? We propose that they are in cases where ignorance of CRNA is implausible, due to the nature of the patient’s condition and clinical context, or because of the objectively high cost of the treatment, or because the patient has disclosed CRNA.

To illustrate, consider physicians who work at the cutting edge of cancer care and have access to innovative medicines that cost hundreds of thousands of dollars per course of treatment. In this situation, it would be reasonable to expect them to recognize cost as a likely major barrier to accessing treatment, and that in cases where patients do not have insurance coverage, they will probably struggle to afford innovative treatment. This will apply for any specialty or condition where the use of high-cost medicines is common. Under these circumstances, CRNA is highly predictable, and we would therefore argue it is reasonable to expect physicians to be aware of this risk and to attribute blame to them where they are heedless of this risk. We can also imagine situations in which there is no reason to believe CRNA is likely, yet individual patients divulge that they cannot afford prescribed medication. For instance, a physician may prescribe a medicine that is subsidized by the government and therefore widely affordable, but a patient may not be eligible for subsidy due to not meeting certain criteria (e.g. they are not legal residents), and informs their doctor of this fact. In cases like this, we propose that physicians are responsible for acknowledging and responding to the patient’s circumstances. On the other hand, the duty is limited in cases that differ from this one. For example, where the medicines are not especially expensive, there is no reason to believe CRNA is likely, or the patient does not disclose their financial situation despite the physician providing them with the opportunity to do so. In such cases the physician would not meet the epistemic conditions for blameworthiness, as they could not reasonably be expected to have the relevant knowledge.¹⁰

This takes us to the second question. To what extent are physicians responsible for knowing about the costs and effectiveness of alternative treatments? It should be noted at the onset that it is not reasonable to expect physicians to be health economists; evaluating the relative cost-effectiveness of different treatments in any rigorous way is a challenging exercise and requires expert economic analysis, and even then there are significant uncertainties involved. Indeed, entire government agencies in many countries are tasked with trying to accurately predict the cost-effectiveness of treatments. However, we do not need to make such strong demands of physicians. Rarely is there only one effective treatment for a condition, and we can expect physicians to be aware of at least some affordable alternatives relevant to their scope of practice. Medical treatments evolve over time and are superseded by better and typically more expensive treatments, meaning that older and cheaper but still effective medications are often still available. To deny that physicians have a duty to be aware of such cheaper options would amount to requiring them to know only of the newest, most effective, and typically most expensive medicines available.¹¹ This would also effectively limit access to many medicines to those who are most wealthy, and deny treatment to the most vulnerable in society, thereby compounding and exacerbating systemic inequity (Ghinea 2022a).

However, there are other types of knowledge that would be unreasonable to demand that physicians possess. For instance, if the only feasible option for addressing CRNA was for patients to import cheaper medicines from abroad via unregulated supply chains, questions of product quality arise. It would be beyond the scope of a doctor’s expertise to know whether such products are reliable. In these cases, we may withhold blame from physicians for not suggesting these alternative options, even if they were aware of them. Alternatively, it could be argued that physicians should still advise patients of these options to the extent that they are aware of them, while providing full disclosures regarding potential risk and uncertainties, if they believe it could be in patients’ interests. Such advice could be specific—in the sense of directing patients to specific suppliers—or general—in the sense of advising patients they may be able to find cheaper alternatives abroad. It is worth noting that a recent study from Australia found that among patients who imported medicines, physicians were the primary source of information about online suppliers (Ghinea et al. 2023). This would be especially relevant in cases where the likely benefits outweigh the harms (e.g. patients with serious illness and no other affordable treatment options).

Given the complexity of insurance systems, it would also be unreasonable to demand that physicians know in detail the out-of-pocket costs of alternative treatments for different patients. Costs can vary widely between patients and treatment options, based for example on the patient’s insurer, type of insurance, government subsidies and where the patient is located. To expect physicians to be across the vicissitudes of health insurance would be unreasonable. It is worth noting that steps are being taken to address this, with the American Medical Association advocating for the development of real-time prescription benefit tools that can be used at the point-of-care to determine out-of-pocket costs for patients, in order to improve cost-transparency and medication adherence (Lubell 2021).¹²

In summary, we believe it may be unreasonable to expect physicians to determine in every clinical encounter what “affordable” means for each patient. For this reason, we propose that physicians are warranted in assuming that patients are not experiencing CRNA, unless there are good grounds for believing otherwise as per the examples we have provided. In cases where CRNA is a foreseeable risk, physicians have additional epistemic duties to know about affordable alternative treatments relevant to their scope of practice.¹³

Control Grounds

Another crucial factor to consider in delineating the scope of physicians’ responsibilities is the extent to which they have control over the circumstances that may impact patient outcomes. The outcome of any medical treatment is inherently uncertain, and for that reason holding physicians to account for the occurrence of rare but nevertheless known side effects would be unreasonable. In contrast, advising patients of these uncertainties is not merely reasonable but required since imparting this information is something that physicians can control. Similarly, physicians cannot force patients to comply with their advice; if a patient does not wish to cooperate, a physician should not be held responsible for the consequences. Other circumstantial factors may also impact the options available to a physician and their patients. To give an example, a patient may have a condition for which a very promising treatment is currently undergoing late-phase clinical trials, but the patient may not meet the eligibility criteria to enroll in the trial, or the trial site may be geographically inaccessible for them. It seems reasonable that a physician be exempted from blame in this case, given that they have no control over the conditions rendering the treatments inaccessible to the patient.

While it is broadly recognized that a physician is accountable for harms arising from failing to disclose information about the advantages and disadvantages of treatments (i.e. failure to gain informed consent), or from failure to maintain relevant skills and knowledge to practice medicine to an accepted professional standard (Ghinea 2023), this duty is limited by practical considerations. Physicians who have breached a duty may be absolved of blame by courts if taking the required action to fulfill that duty is deemed unreasonably burdensome or the failure can be justified on other grounds. For instance, as Wendler claims, a physician may be absolved of blame for failing to single-mindedly serve an individual patient’s best interests if doing so would undermine the care of other patients for whom they are also responsible (Wendler 2010). This reasonableness criterion is effectively a control condition limiting a physician’s accountability for their actions.

Another way of thinking about control conditions proposed recently by Ghinea is to think of a patients’ financial circumstances as an intervening cause of harm, and therefore not one that can be used to attribute blame (Ghinea 2023). For instance, failing to provide patients with adequate information about treatment options, or failing to ensure one’s skills and knowledge are current, are failures that rest with the physician and for which they can be blamed, because it is within their capacity to control the circumstances that give rise to the failures. However, CRNA does not reflect a failure on the physician’s part since the causes that make a patient unable to afford treatment are not within their control. If physicians were to be held accountable for harms that do not arise from their own actions it would be difficult to see where their responsibilities would end.¹⁴ According to this framing, claims Ghinea, it is plausible to conclude that physicians only have a responsibility for those harms the root causes of which they have some degree of control over.

It would also be unreasonable to demand physicians to act in ways that would expose themselves to a substantial risk of liability, even if it may be in the interests of their patients. For instance, during the controversy around the off-label prescribing of bevacizumab for aged-related macular degeneration in the United Kingdom, physicians were advised by the General Medical Council that such prescribing was illegal (Cohen 2015; Dickson 2015). This deterred many physicians from prescribing this much cheaper alternative to the on-label medicine ranibizumab. However, many physicians did continue to prescribe bevacizumab, but such advocacy in light of legal ambiguity would be best construed as a professional aspiration rather than obligation.

Combining these considerations regarding the control condition, we propose that physicians do have a responsibility to, at the very least, disclose information about affordable alternative treatments of which they are already aware (or ought to be aware given their scope of practice), if doing so would avert harm to their patients. This is because the physician is asked to do nothing more than share information they already possess or ought to possess, which is the least burdensome task that can be demanded of a physician, and something they have proximate control over.¹⁵

STRATEGIES FOR BYPASSING COST-BARRIERS TO MEDICATION ADHERENCE

The preceding arguments have sought to delineate the grounds and limits for the proposal that physician duties can and should encompass mitigation strategies for CRNA. At the minimum, this duty involves informing patients at risk of CRNA of any more affordable alternatives. However, claiming such a duty is pointless if there are no options for fulfilling it. Here we illustrate a number of strategies that are already utilized to help overcome cost-barriers to care, some of which are more controversial than others, and highlight the challenges associated with each (Ghinea 2022a).¹⁶ The existence of these strategies provides implicit support to our view that it is part of physicians’ duties to address CRNA to prescribed medication.

Perhaps the most important method of addressing cost is to substitute generic or biosimilar drugs for the brand name versions.¹⁷ Such products are substantially cheaper than their brand name counterparts, but they are usually only legally available once relevant patents expire, which can take many years. This means not all medicines have generic or biosimilar competitors in the local market, and indeed, in some cases, these copies may still be too expensive for some patients to afford.

A second method to reduce medication costs is to enroll patients in clinical trials, in which medications are supplied free of charge to the patient. This strategy is relatively common for gaining access to high-cost treatment in specialties such as oncology. However, this assumes the existence of suitable trials for which a patient is eligible, that patients and doctors are aware of such trials, and that extraneous expenses do not make participation unfeasible (e.g. travel expenses or additional unfunded testing).

Third, pharmaceutical companies may offer individual patients discounted or free access to specific medicines upon request. However, fulfillment of such requests is ad hoc and driven by industry’s commercial interests. Hospitals similarly may consider requests to fund treatment, but again the decision is at the discretion of drug committees or the hospital executive. These strategies for reducing medication costs are commonly deployed and generally not controversial.

More controversial strategies for addressing CRNA to medications have the potential to significantly impact the pharmaceutical industry’s commercial interests. The first is the use of cheaper off-label products in preference to on-label alternatives. This is illegal in some European jurisdictions even if the off-label use offers the same degree of safety and efficacy as available on-label alternatives (Raposo 2017). In the United Kingdom, the General Medical Council had advised that doctors would be in breach of duty if they prescribed off-label medicines when on-label alternatives were available (Dickson 2015). However, UK courts have recently ruled in favor of using off-label medicines for cost reasons, at least in some cases (Dyer 2020). Nevertheless, there remains significant ambiguity about the legality of this practice in various jurisdictions, a fact that may itself deter physicians.

A second controversial strategy to reduce costs is through importation of cheaper medicines from abroad. This is particularly controversial in cases where importation is used to access generic and biosimilar products prior to patent expiry in the patient’s country of residence. There is a flourishing international online marketplace for medicines. These markets and their supports occur in various forms, such as buyer clubs like FixHepC which provide affordable access to specific medicines for particular patient populations (Ghinea, Kerridge, et al. 2017); medicine intermediaries like Everyone.com that help patients access medicines not available in their countries; commercial comparison websites like PharmacyChecker.com that provide patients with price comparisons across international pharmacies verified for safety and efficacy; and online retail websites like IndiaMart that connect consumers with a host of medicine suppliers willing to export internationally. However, the European Court of Justice has condemned economically motivated prescribing in cases where an approved product with the same active substances, dosage, and form exists in the market (Raposo 2017). On this ground, the Polish government has been prosecuted for allowing medicines to be imported for cost reasons (Raposo 2017).

It is worth noting that in the United States personal importation of cheaper medicines has been the subject of extensive lobbying and debate for over a decade due to the high prices in the US market. In most circumstances it is illegal, but there are situations in which it “might” be permitted and a thriving market in personal imports seems to exist (US Food and Drug Administration 2023). In some European countries, personal importation is permitted, but only from specific regions. For instance, Finland permits importation from countries belonging to the European Union or European Economic Area (Buying medicines online 2023). The United Kingdom has no official position on personal importation, however, advice from advocates suggests that it is permitted (Ghinea 2022b). In countries including Australia, Japan, Switzerland and New Zealand importation of medications for personal use is permitted with certain caveats (Ghinea 2022b).

The biggest concern associated with importation relates to the safety and quality of medicines sourced through unregulated supply chains (McManus and Naughton 2020). Doctors helping patients import medicines, or who are managing patients who use imported medicines, need to consider the potential risks and benefits of doing so. However, given that importation may be the only option available for some patients to access affordable treatment, assisting patients may be the least worst option when compared with refusal to assist (Ghinea 2022a).

CONCLUSION

Evidence shows that CRNA to prescribed medicines is common and widespread, and that it adversely impacts the health and wellbeing of patients. We have argued that physicians cannot ignore CRNA, due to the responsibility they have to protect patient health and wellbeing. Since the realization of health benefits cannot be decoupled from access to treatment, we have proposed that physicians have a responsibility to prescribe not only effective treatment but also treatment that their patients can reasonably be expected to afford.

We have also argued that since medical professionals and governments have demonstrated a willingness to accept higher risks when treating unmet health needs and neglected health conditions, they should also be willing to address CRNA, on the grounds that there is no moral or practical difference—for the patient—between medicines being unavailable and medicines being unaffordable. This can include such practices as prescribing cheaper off-label treatments and supporting patients to import more affordable medicines from abroad.

Given the significant impact of medicinal CRNA, explicit guidance will need to be developed and provided to enable and support physicians to navigate their responsibilities for affordability-motivated prescribing. It is our hope that this article yields the conceptual groundwork and principles on which such guidance can be based.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was made possible by a Macquarie University Research Fellowship granted to Narcyz Ghinea.

Notes

1 It is important to note that in the United States the cost of insulin for non-Medicare enrolees is still high, and other aspects of diabetes care remain expensive so access to affordable treatment remains a major challenge.

2 Off-label prescribing refers to the use of a licensed medicine in unlicensed ways. This is legally permissible as therapeutic goods regulators do not have authority to interfere with a physician’s professional autonomy when it comes to prescribing treatments.

3 Throughout we mean health and wellbeing in the clinical sense of treating or managing disease.

4 For instance, is the fact that 10% of Australians in poor or fair health report CRNA sufficient to recognize this as a significant health-related risk factor that should trigger doctors to take remediating steps?.

5 Gruen et al. do recognize that control conditions are important in delineating the limits of a doctor’s duties. We address control conditions later in the paper.

6 Of course, newer medicines may not always be more effective for specific patients due to safety profile, drug interactions, and other reasons. However, the general trend is that new medicines are more effective overall (even if only incrementally so) and more expensive.

7 Of course, this would not absolve a physician who prescribes a lower-cost medicine that is inferior to the gold-standard treatment from their duty to provide adequate disclosure. Otherwise, there is a risk of physicians’ presumptions about patients’ financial circumstances leading to mis-prescribing.

8 We focus on physicians as while other health practitioners such as nurse practitioners may have prescribing rights, they generally lack the same degree of professional discretion as physicians when it comes to making clinical judgements that demand balancing risks versus benefits.

9 Other health professionals with prescribing rights may bear this burden to the degree that corresponds to their scope of practice.

10 In cases where patients are taking multiple medicines, it is reasonable to extend this reasoning to the cumulative cost of medicines.

11 While the launch prices of new medicines in the United States continue to increase, reaching a median price of around $200,000 per annum in 2022, it is worth noting the relationship between price and years on the market is not straightforward.

12 However, we still propose that physicians have a duty to have a general sense of what treatments are generally cheaper than others as described earlier.

13 We argue in the next section that reasonability criterion applies to this epistemic duty.

14 This is potentially an alternative way of interpreting Gruen et al’s distinction between professional aspirations and duties discussed earlier. It should be noted that Ghinea sees this as a plausible argument, but not one that he defends.

15 We are effectively proposing that physicians have a duty to gain informed financial consent, not merely informed consent in the traditional sense. By this we mean physicians offer patients options based not only on effectiveness, but also affordability.

16 In addition to the general mechanisms we describe here, specific jurisdictions may have other mechanisms available to help patients access more affordable medicines such as national or state grants (e.g. US’s Ryan White HIV/AIDS Program Legislation) or drug price checking websites and coupons (e.g. GoodRx).

17 The term “generic” is used to describe copies of chemically synthesized drugs, whereas “biosimilars” refer to copies of biologics which are made from, or derived from, living cells or organisms.