• complexity
• compliance
• health services research
• primary care

It is usually a grand affair when ‘Ms Noelle’ makes it to clinic. The 52-year-old mother with a history of hepatitis C cirrhosis, hypertension, uterine fibroids and migraines has been in our care for over a year. Even so, each visit still brings a new crisis. Today, we found out that Ms Noelle, the caretaker of a daughter with bipolar disorder and nine grandchildren, had just been evicted from her home. She had been without any income for months, and her applications for temporary cash assistance and disability were denied. Ms Noelle maintained a remarkable ability to keep her family protected and fed despite all this, but we have watched as she became the ultimate victim: she struggled to remember her medications, their doses and indications, and her cirrhosis was frequently on the verge of decompensation during appointments she was barely able to keep. She was overwhelmed by even the simplest of tasks, such as completing forms or picking up meds from the in-house pharmacy, despite how much she knew she needed to follow through.

We make innumerable health choices daily. Behavioural economics can help us understand these choices, as well as the logical errors to which all patients are vulnerable. For example, patients may downgrade the importance of future health status in relation to current gratification.1 Individuals are also risk averse in decision-making, valuing losses disproportionally relative to gains, which makes them susceptible to make vastly different treatment decisions dependent on how a choice is presented.1 Further, decisions made in certain emotional states may not translate to behaviour change when in different emotional states (eg, committing to a diet while in a calm state, then succumbing to dessert when stressed).2 These relatively small choices and errors may have substantial effects on health behaviours and health outcomes.

Patients who have multiple chronic conditions and live in poverty, such as Ms Noelle, are constantly negotiating a number of trade-offs created by a lack of economic resources. As a result, they are disproportionately affected by cognitive errors. The ‘mental bookkeeping’ required to survive under these conditions might actually lead them to develop heightened accounting skills compared with higher income counterparts.3 However, under chronic scarcity, the energy put towards survival overwhelms natural mental capacity (a phenomenon described as ‘attentional capture’), reducing fluid intelligence, the mental capacity required for other tasks.3 The ‘cognitive tax’ brought by the stress of navigating chronic scarcity has been measured to be equivalent to a drop of 13 IQ points—enough to push individuals from ‘average’ intelligence to ‘borderline-deficient’—thus increasing susceptibility to commit cognitive errors in activities that fall out of the sphere of pressing concerns.4

Poverty-related attentional capture, while shown to be reversible, has been cited as an contributing factor behind so-called ‘paradoxes’ such as low uptake of social services and dependency on high-interest loans.3 Attentional capture affects how low-income individuals engage with complex health interventions and health systems, as managing the myriad demands that must be fulfilled to survive in low resource settings—financial and non-financial in nature—can limit retrospective and prospective memory necessary for patients to engage in healthcare. This has implications for adherence, as forgetfulness is one of the most cited causes of non-adherence.5 For example, in a study of callers to a United Way assistance hotline, researchers demonstrated that individuals with increased levels of stress related to unmet basic needs such as housing, food or safety, displayed diminished recall of provided referrals for cancer prevention services, leading to reduced levels of follow-up.6

Higher levels of therapeutic burden have been shown to relate to lapses in engagement in patients of all socioeconomic strata.5 7 A recent perspective demonstrated that a theoretical patient with non-insulin-dependent diabetes, hyperlipidaemia and hypertension has to complete at least 12 health-related actions daily and 4563 annually to adhere to standard recommendations of health maintenance.8 Compound these tasks with the unique costs that low-income patients might bear to reach medical care—such as financing transportation, missing work and funding child care—and it follows that chronic disease self-management and care coordination might become a disproportionate burden for those experiencing chronic financial stress.

We should strive towards ‘minimally disruptive medicine,’ tailoring treatment regimens and delivering care that minimises burden on patients.7 Opportunities to alleviate undue cognitive burden exist at the provider, practice, health system and community levels.

At the provider level, recognition of poverty-related stressors should guide clinician–patient discussions. Identifying and aligning priorities to reflect patients’ most potent life stressors in initial encounters is important to building a therapeutic relationship, even if those stressors are non-medical. In an analysis of medical system ‘home runs,’ Milstein and Gilbertson identify success as dependent on reciprocation, claiming that programmes that improved outcomes tended to engage patients in their own care by convincing them that their overall well-being, not just health outcomes, was the provider’s chief concern.9 Using motivational interviewing to build self-efficacy in sustaining behavioural change should also be taken into account, particularly as self-efficacy has been shown to be eroded by multidimensional poverty.10 Clinical teams should partner with patients to achieve small ‘wins’ in financial and health challenges, thereby increasing patients’ confidence and eventually improving adherence.11

For providers, simplifying healthcare also involves reducing complex regimens when possible, synchronising refill dates across medications, and prescribing in 90-day cycles to reduce the risk of non-adherence.12 Fixed-dose combination medications, despite their dosing limitations, might be a useful option, and prudent deprescribing where possible is a must.13

Appointment adherence is a potential area of intervention at the practice level. In 2010, Americans made 1.03 billion visits and spent 2.4 billion hours seeking outpatient care for themselves or loved ones, 31% of which was spent travelling to appointments and only 17% was spent face-to-face with physicians, with the rest spent waiting to ‘check-in’, ‘check-out’ or see a provider.14 Racial minorities and unemployed patients spend even more time travelling and waiting for care with no increase in face-to-face time with providers.15 A combination of practitioner maldistribution and shortages in underserved areas as well as practice-level workflows contribute to this time disparity.16 17

Care must be organised efficiently from the patient’s point of view; otherwise, the myriad of competing priorities that underserved populations face will deter access to appropriate outpatient care. One potential solution is to provide flexible scheduling for patients. Open-access scheduling and extended hours may provide greater latitude to low-income patients who might be otherwise forced to make trade-offs on their healthcare for child care or their work. Data demonstrate that open-access scheduling may decrease no-show rates, and extended hours paradoxically decrease healthcare expenditures.18 19

How traumatised patients experience  clinical care is also an important practice-level consideration. Chronic socioeconomic stress, often coupled with childhood traumas and ongoing social and family instability, increases the risk of chronic medical and psychiatric diseases.20 Health clinics may trigger a strong emotional and behavioural reaction due to perceived lack of control and emotional and physical exposure inherent to medical settings, creating a situation in which adults with a high disease burden paradoxically avoid healthcare.21 Trauma-informed care principles should be incorporated across a practice to create a caring environment for patients. Engaging patients in a physically, psychologically and emotionally safe space will help them manage stressors in and out of the clinic.21

Beyond providers and individual practices, delivery innovations may limit the number of cognitive steps needed to use and access healthcare systems. Mobile technology has great potential to engage patients in care, to support chronic disease self-management and to promote behaviour change, such as with medication adherence or smoking cessation.22 While there is limited evidence for use in low-income populations, there are ongoing studies in this area.23 To expand access outside the clinic, home visits by paraprofessional programmes have demonstrated improvement in adult asthmatic control and in child health in low-income settings.24 25

Direct financial assistance in the form of conditional cash transfers has also led to improved health system engagement in low-income and middle-income countries. A systematic review on the topic demonstrated increases in individuals returning for voluntary HIV counselling and number of children taken to health facilities, although data on immunisation uptake and health outcomes are mixed.26 Also, a conditional cash transfer programme in New York City decreased poverty and improved adult and child health status.27 Further research on how this tool may engage US patients, target health behaviours and affect health outcomes is needed.

Finally, it is most important to emphasise that cognitive costs brought about by the stress of navigating chronic scarcity have been shown to be reversible when financial constraints are ameliorated,3 which points to the need for community-level interventions that liberate mental ‘bandwidth’ for patients to focus on long-term health goals. The study of United Way callers demonstrated higher odds of follow-up in individuals who had all their material needs addressed after the original screening.28 Various nationwide models of integrated care have focused on partnerships with community organisations to link them with patients with material needs. One such programme, Health Leads, has demonstrated gains in low-density lipoprotein and blood pressure control among clients with satisfied social needs.29 The Center for Medicare and Medicaid Innovation’s Accountable Health Communities model is another example of community partnerships.30

While many of these ideas have been piloted or proven in other contexts, some might require experimentation to demonstrate effectiveness in low-income populations. Furthermore, the proposals might need to be adapted to contexts in low-income and middle-income countries where resources may be limited, although the need to acknowledge scarcity-induced distress might be even more pressing. If we are to make progress at improving health outcomes for our most vulnerable patients, we must reimagine and reshape clinical interactions, as well as the structure and function of our health system and communities. Only then may we simplify the treatment burden many of our patients face.

Ms Noelle remains challenging to engage, although her son now helps connect her to care. In conjunction with a behavioural specialist, we created a treatment plan to prioritise her medical, behavioural and social needs. She committed to a hepatology appointment once her income stabilises. We move towards goals incrementally and celebrate all achievements, including coming to clinic appointments. And most of all, we strive to provide compassionate care that preserves Ms Noelle’s dignity and empowers her to progress forward.