Should doctors record their patients’ income?

INTRODUCTION

The link between lower income and poorer health is well established. Lower income is known to be associated with lower life expectancy and higher rates of heart disease, cancer, and mental illness.¹ Access to health care is also known to be poorer for people with lower incomes. Primary care is frequently the point of first contact between healthcare services and individuals with health and social problems. Yet, income data is not routinely collected in primary care. Knowledge of patients’ income is consequently not incorporated into the clinical care of individuals, and is underutilised in policy making and healthcare planning for populations. Income interacts with behaviour, actions, and environment to impact health across the life course and across all sections of society. Evidencing, understanding, and acknowledging these interactions is essential if we are to tackle inequities in health.

Should doctors in primary care record their patients’ income? We argue that it would bring individual and population benefits; that acceptability and practical applicability may be less problematic than first supposed; that a precedent exists in the routine collection of other sociodemographic data; and that the UK is lagging behind other countries in considering this issue.

PATIENT AND POPULATION BENEFITS

Household (rather than individual) income would be the most useful data to record in most circumstances. Benefits of recording patients’ household income can be anticipated at a number of levels (Box 1). For patients, the quality of health care received may be improved if their doctor is aware of their household income. Doctors would be better able to: offer health and lifestyle advice suited to a person’s budget; identify patients for whom prescription charges (or other healthcare-related costs) may represent a challenge to compliance; and ensure patients are accessing community resources that could be helpful to them and state benefits to which they may be entitled.²

Box 1. Anticipated benefits and barriers to recording patients’ household income.

Benefits	Barriers
Patients
Improved health and lifestyle advice, better-suited to individual’s budget Improved potential for discussion with doctor about underlying causes of disease and illness Improved access to health and social care interventions to alleviate poverty and promote health	Concerns over perceived loss of privacy Concerns about confidentiality (such as, that information may be shared) Concerns that healthcare attitudes or actual care provision may be adversely influenced by knowledge of their income Concerns that income inquiries may be used to check up on entitlements to state benefits Certain groups may be less willing to disclose income Some patients may not know their overall household income
Healthcare providers and health systems
Increased attention on the household and environment causative factors in disease Role of poverty as a potential determinant of illness emphasised during clinical encounters and More rational health system structuring to provide services that meet patients’ needs collection media	Concerns that income enquiries may impair doctor–patient relationships Concerns that primary care doctors will eventually be tasked with screening for administering access to state benefits Concerns about how income data will be portrayed in the Concerns about intrusion in patients’ private lives and/or storage of sensitive patient data Concern that income data may be used to support restructuring of health services, contributing to a ‘postcode lottery’
Public health and academic
Greater understanding of relationships between income and healthcare access, healthcare experience, and health outcomes, Greater understanding of health inequities and their assessment Improved commissioning to support delivery of health and social care services aimed at alleviating poverty and illness, and promoting health	Concerns about media portrayal Concerns about accuracy and adequacy of data for strategic decision making

In UK primary care, one of the few means by which socioeconomic data is currently incorporated into clinical management is by means of QRISK: the prediction algorithm commonly used to calculate cardiovascular disease risk and guide therapeutic decision making. QRISK calculations incorporate traditional risk factors (including age, sex, blood pressure, and smoking status) plus the Townsend Deprivation Score derived from the individual’s UK postcode.³ Townsend Scores are based on census data collected every 10 years: arguably, as indicators of deprivation they are inadequately responsive to today’s rapid demographic changes. Furthermore, like any postcode-based indicator of deprivation (including the Index of Multiple Deprivation, often used in health research) the Townsend Score is unreliable where local populations are heterogeneous, is invalid for mobile populations or the homeless, and is liable to confounding by the area effect (postcodes may be associated with levels of deprivation, but may also be independently associated with other environmental health determinants, such as pollution). Through our unquestioning use of such flawed metrics we may be doing a disservice to our patients; particularly some of those who are already most vulnerable and socially excluded. Household income might ultimately offer an alternative or additional variable.

Linking household income to healthcare data would be of use to researchers and public health planners. It would enable better recognition of income-associated variation in healthcare access, healthcare experience, and health outcomes; permitting better-informed and more effectual corrective interventions. It could reveal the extent to which (if at all) associations between low income and poorer health are ameliorated by state benefits in Britain and it could help to disentangle the various confounded measures of deprivation.⁴ If we are to take seriously the issues of income and health inequities in Britain we need reliable knowledge of each, and of their interactions.

SHOULD DOCTORS DIAGNOSE POVERTY?

A strategy of ‘screening for poverty’ in primary care has been proposed in Canada.⁵ Canadian researchers found that the enquiry ‘Do you (ever) have difficulty making ends meet at the end of the month?’ had a sensitivity of 98% in detecting people below a determined ‘low-income cut-off’ or ‘poverty line’.⁶ In Australia, a Health Ministry summit recently heard that ‘GPs should routinely screen patients for poverty to cut deaths from preventable diseases’.⁷

A label of ‘poverty’ could be stigmatising for individuals. However, poverty itself arguably bestows more stigma than any label, and more health risk than any concern about stigma could outweigh. Unlike diagnostic thresholds for other risk factors, the ‘low-income cut-off’ would not be arbitrary: UK researchers have already defined a theoretical ‘Minimum Income for Healthy Living’ (and demonstrated that state benefits and minimum wage payments in the UK fall far short of this threshold).^8,9 These theoretical calculations could be usefully supplemented by evidence on income and health from primary care to provide a strengthened basis for discussion on public policy and standards of social provision. In the UK, where income and health inequalities are worse than in either Canada or Australia, and among the worst in Europe,¹⁰ discussion and action on these issues is urgently required.

Poverty screening complements enquiries about household income but household income alone may not provide a complete picture of an individual’s available resources. Meanwhile, income-associated health inequities affect everyone, from the top to the bottom of society; not only those below the poverty line.¹ So both poverty screening and income enquiry could be valuably included in a consultation or annual review.

The suggestion of documenting income and diagnosing poverty may prompt accusations of unwarranted medicalisation; but income and poverty become medical issues when, as is inevitably the case, they impact on a person’s health. Furthermore, knowingly medicalising an issue need not be a bad thing if it brings some improvement to people’s condition through better research and understanding of health inequities; better targeting of interventions; and some political motivating and advocacy around income inequities and their health effects. These endeavours are manifestly the job of the health professions.¹¹

WOULD PATIENTS REVEAL THEIR INCOME TO DOCTORS?

Patients would not necessarily have to reveal their actual household income. Although exact household income would provide the best measurement of wealth status, in practice, income brackets could be used, as in social surveys: in UK social surveys, income data derived from a single question has been shown to be generally accurate, with a response rate of 85–95%.¹²

Certain groups may be unwilling or unable to reveal their household income; for example, due to suspicions regarding the purpose of income-related questions, or lack of awareness of their overall household income. Patients and advocacy groups may also be wary about data security and breaches in confidentiality. Disclosure would not be obligatory, but explaining to patients why the information was being requested and how the data could be used, as well as being clear about methods for managing risk and confidentiality, should help to alleviate concerns. Ultimately, the income information that our patients readily provide to bank managers and mortgage advisers could also be usefully imparted to doctors.

A precedent is found in the recording of ethnicity in UK primary care. Before ethnicity became routinely documented in clinical records a thorough analysis of feasibility and acceptability to both staff and patients was conducted.¹³ A similar assessment would need to be undertaken for income data collection. Ethnicity recording also provides an exemplar for the logic of income data collection: the practice was financially incentivised on the grounds of improving research capability and the targeting of resources.¹⁴ From a medical ethics perspective, the recording of patients’ income may be more ethically acceptable than recording ethnicity, since patients are better able to exercise their autonomy by refusing to disclose income information for documentation if they wish.

CONCLUSION

Should we record our patients’ household income? And be prepared to diagnose poverty? It may be a discomforting prospect. But our discomfort is poor justification for not recognising the evolving role of general practice; for avoiding truths about the extent of health inequities in Britain; for failing to acknowledge the lives blighted by these inequities that many of our patients lead; and failing to respond in clinical care and public health planning. It would be shameful indeed if our sensitivities about discussing income with individuals in primary care constituted yet another means by which health inequities were permitted to quietly continue increasing.

Provenance

Freely submitted; not externally peer reviewed.