Elsevier

Journal of Health Economics

Volume 50, December 2016, Pages 372-387
Journal of Health Economics

Patient mobility and health care quality when regions and patients differ in income

https://doi.org/10.1016/j.jhealeco.2016.05.003Get rights and content

Abstract

We study the effects of cross-border patient mobility on health care quality and welfare when income varies across and within regions. We use a Salop model with a high-, middle-, and low-income region. In each region, a policy maker chooses health care quality to maximise the utility of its residents when health care costs are financed by general income taxation. In equilibrium, regions with higher income offer better quality, which creates an incentive for patient mobility from lower- to higher-income regions. Assuming a prospective payment scheme based on DRG-pricing, we find that lower non-monetary (administrative) mobility costs have (i) no effect on quality or welfare in the high-income region; (ii) a negative effect on quality but a positive effect on welfare for the middle-income region; and (iii) ambiguous effects on quality and welfare for the low-income region. Lower monetary mobility costs (copayments) might reduce welfare in both the middle- and low-income region. Thus, health policies that stimulate cross-border patient mobility can be counterproductive when regions differ in income.

Introduction

Cross-border patient mobility is currently a key issue for health policy. In the European Union (EU), patient mobility across member states has been high on the political agenda for many years, despite the fact that the free movement principles do not apply to health care provision. A key example is the new directive adopted by the European Parliament and the Council in 2011, which gives patients the right to choose among health care providers across all EU member states.1 On the 25th of October 2013, when the directive came into force, the Health Commissioner Tonio Borg said2:

“Today is an important day for patients across the European Union. As of today, EU law in force enshrines citizens’ right to go to another EU country for treatment and get reimbursed for it (…). For patients, this Directive means empowerment: greater choice of healthcare, more information, easier recognition of prescriptions across-borders.”

The actual patient flows across EU member states are small. In 2011, when the new directive was implemented, the European Commission estimated the demand for cross-border health care to € 10 billion, i.e., 1% of public health-care spending in the EU. In a recentEurobarometer survey, only 5% of the respondents reported that they had received medical treatment in another EU country.3 However, according to the same survey, 49% said they would be willing to travel to another EU country to receive medical treatment with the two main reasons being to receive treatment not available in home country (71%) and to receive better quality treatment (53%). These figures suggest a large potential for cross-border health care within the EU.

Cross-border patient mobility is also an important policy issue for countries with regional health-care provision. Sweden implemented a ‘free choice’ reform in 2003 that allowed patients to demand health care outside their home county and ensured cost reimbursement by specifying transfer payments for cross-border care. A similar system is in place in Italy, where many patients migrate from the south to the north to obtain better medical care. However, in Canada, patient mobility across provinces is generally limited to emergency and sudden illness or allowed only in special circumstances. In the US, state-specific regulations restrict individuals from purchasing health insurance outside their home state, which limits patient mobility across state lines.4

These real-world observations highlight the importance of understanding the effects of cross-border patient mobility. Would the new EU directive improve access to better health care for patients, or is the reform beneficial only for some countries while others are worse off? Would cross-border patient mobility in the EU stimulate quality provision among member states, or can the reform have adverse effects on the quality of care? Should Canada do as Sweden and Italy by allowing for patient mobility across regions, or are they better off with the current system?

In this paper, we study the impact of cross-border patient mobility on health-care quality and welfare, and discuss the implications for health policy. To do so, we develop a spatial model á la Salop (1979) with three regions that differ in income distribution, i.e., a high-income, a middle-income, and a low-income region. In the context of EU, we can interpret the high-income region as Northern European countries, the middle-income region as the Southern European countries, and the low-income region as the new member states in Eastern Europe.

The policy maker in each region chooses quality to maximise the utility of its own residents subject to a budget constraint, where the total cost of health care provision is financed by general income taxation. To allow for income effects, we assume individuals have decreasing marginal utility of income, which implies that the marginal cost of raising tax revenues decreases with average income. Consequently, health care quality is increasing in the regions’ income level, inducing cross-border patient flows from poorer to richer regions. We focus on the equilibrium where the high-income region only imports patients, the low-income region only exports patients, whereas the middle-income region both exports and imports patients.

In the analysis, a key assumption is that the regions apply a prospective payment scheme where the health care providers receive a fixed price equal to the treatment cost per patient. This payment scheme is consistent with Diagnosis-Related Group (DRG) pricing that has been adopted by almost every European country, and is now by far the most commonly used payment scheme in the EU (Busse et al., 2011). As the treatment costs are arguably higher in richer regions with better health care quality, DRG-pricing implies that the exporting regions are facing a net financial loss related to cross-border patient mobility, whereas the importing regions are fully compensated for the treatment of migrating patients. Another key assumption is that we allow for the exporting regions to possibly charge a copayment from the migrating patients to cover the difference in the treatment costs. This assumption is in line with the new EU directive that entitles patients that demand cross-border care within the EU to cost reimbursement equal to that of their home country.5

Based on this framework, we derive a rich set of results regarding the regional effects of cross-border patient mobility on quality provision and welfare. First, a reduction in non-monetary mobility costs (e.g., a simplification of administrative procedures) has no quality or welfare effects in the high-income region, since the costs of treating migrating patients are fully compensated. However, for the two other regions, a reduction in non-monetary mobility costs tends to reduce quality. In both the middle-income and the low-income regions, the direct effects of lower (non-monetary) mobility costs on quality provision are unambiguously negative. Increased patient mobility reduces the marginal benefit of quality provision, because fewer patients will be treated in their home region. In addition, since patient export has monetary costs – both for the patients who migrate and for the tax payers – higher mobility needs to be financed by a higher income tax rate for exporting regions. For the low-income region, we also identify an indirect effect related to quality provision in this region being a strategic substitute to the quality provision in the middle-income region, which counteracts the direct effects mentioned above.

Second, the effects of reducing monetary costs of patient mobility (i.e., patient copayments) are qualitatively similar to the effects of reducing non-monetary costs. However, there is an additional budget effect that makes the overall effect generally indeterminate. A lower copayment implies that a larger share of the costs of patient export needs to be financed by the exporting regions’ tax payers, which in turn implies a tightening of the government's budget constraint. This gives the exporting (low- and middle-income) regions an incentive to increase quality in order to mitigate the increase in mobility caused by lower patient copayments.

Third, the effects of cross-border health care on regional welfare are mixed with winners and losers from such a policy. A reduction in non-monetary mobility costs has (i) no welfare effect in the high-income region, (ii) a positive welfare effect in the middle-income region, and (iii) an indeterminate welfare effect in the low-income region. The middle-income region unambiguously benefits because of the cost reduction for the patients who seek treatment in the high-income region. A similar effect also applies to the low-income region. However, in this region there is a potentially counteracting welfare effect due to the quality reduction in the middle-income region, which harms the migrating patients but benefits the tax payers. If, on the other hand, cross-border mobility is stimulated by a reduction in monetary mobility costs (lower copayment), the welfare effects in the middle- and low-income regions are generally ambiguous.

Finally, we analyse the effects of more income dispersion both across and within regions. Larger inter-regional income inequality leads to higher quality in the high-income region, lower quality in the middle-income region, whereas the effect in the low-income region is ambiguous. The effects of higher intra-regional income inequality, on the other hand, depend on the region in which income dispersion increases. Higher income inequality in the high-income region leads to higher quality in that region and lower quality in the other two regions, while higher income inequality in either the middle-income or the low-income region has no effect on quality provision in the high-income region and indeterminate effects on the other two regions. Thus, allowing for cross-border patient mobility can create negative spillover effects of higher income inequality in the form of lower quality of health care in neighbouring regions.

The literature on cross-border patient mobility is limited but growing.6 The recent papers by Andritsos and Tang, 2014, Andritsos and Tang, 2013 use a queueing framework to analyse the effect of cross-border patient mobility on waiting times and reimbursement policies.7 Andritsos and Tang (2013) find that patient mobility can increase patient welfare due to increased access to care. However, the effects on waiting times and reimbursement rates are mixed, and the additional costs of mobility are disproportionately shared between the participating countries. Andritsos and Tang (2014) find that patient mobility can be beneficial to public health-care systems (NHS), as health-care funders can reduce their costs without increasing the patients’ waiting time. In border regions, where the cost of crossing the border is low, ‘outsourcing’ the high-cost country's elective care services to the low-cost country is a viable strategy from which both countries can benefit. Despite similarities, these studies do not consider the effect of patient mobility on health-care quality nor the role of differences in income distribution across and within regions, which is the key focus of our paper.

The closest paper to ours is Brekke et al. (2014b) who consider a Hotelling model with two regions that differ in health-care technology, where the region with more efficient technology offers higher health-care quality and attracts patients from the region with less efficient technology. A key finding is that the effects of patient mobility depend on the transfer payment scheme. If the transfer payment is below marginal cost, mobility leads to a ‘race-to-the-bottom’ in quality and lower welfare in both regions. Thus, patient mobility can have adverse effects on quality provision and welfare unless an appropriate transfer payment scheme is implemented.

In the current paper, we take a different approach by focusing on differences in the income distribution across regions as the key source of cross-border patient mobility.8 This is an issue that has received much attention in the EU debate, especially after the recent extension of the EU to Eastern Europe. To capture this, we therefore apply a model with three (rather than two) regions. This also implies that the same region can be both importing and exporting patients, which cannot arise with the two-region set up in Brekke et al. (2014b). Finally, we use more general cost functions, allow for copayments when patients demand care outside their region, and allow for heterogeneity in income within countries/regions (with richer patients more likely to move). Critically, we introduce income effects through decreasing marginal utility of income, which implies that qualities are in most reasonable scenarios strategic substitutes, and this is an important driver of some key results. We investigate the effect of policy-relevant parameters such as patients’ copayments and inter- and intra-regional income dispersion. Thus, our paper is significantly different from Brekke et al. (2014b).

Our paper also relates to the broader health economics literature on provider competition and quality incentives. A key finding from this literature is that with regulated prices, competition increases health-care quality if providers are profit-maximisers, whereas the relationship between competition and quality is generally ambiguous if providers are (partly) altruistic.9 Despite some similarities, our study differs from this literature as we consider competition between regions (rather than providers), where health-care quality is set by policy makers that maximise regional welfare financed through taxation. Moreover, the income distribution across and within regions is central to our study, but not a part of the previously cited papers.10 Thus, the competitive mechanisms in our model are clearly different from the more general literature on provider competition and quality incentives.11

The rest of the paper is organised as follows. In Section 2 we present our model. Equilibrium quality provision is derived and presented in Section 3, whereas, in Section 4, we describe the strategic relationship between regional quality choices. In Section 5 we analyse the effects of policies to stimulate cross-border patient mobility – a reduction of either monetary or non-monetary mobility costs – on regional quality provision and welfare. In Section 6 we explore the effects of (inter-regional or intra-regional) income inequality on regional quality provision and analyse how these effects depend on cross-border patient mobility. Finally, Section 7 concludes the paper.

Section snippets

Model

Consider a market for health care where patients are uniformly distributed on a circle with circumference equal to 1 and the total patient mass normalised to 1. The market consists of three different regions, which can be interpreted either as neighbouring countries or neighbouring regions within the same country. The three regions, indexed by i = L, M, H, are of equal size, each covering 1/3 of the circle. The index i denotes whether the region has Low, Middle or High average income. The

Optimal quality provision

The policy maker in each region chooses quality to maximise the utility of its own residents subject to a budget constraint, where the total cost of health care provision is financed by general income taxation with a tax rate τi. For patients travelling from Region i to Region j to receive treatment, the provider in Region j receives a payment pj per patient. We assume that part of this payment, π, is paid by the patient himself, whereas the remaining part, pj − π, is paid by the tax payers in

Strategic interaction between regions

In order to understand the main mechanisms involved in the model, it is instructive to study the nature of the strategic interaction between the different regions. In order to facilitate this analysis, and the comparative statics analysis in the subsequent sections, we also make the additional simplifying assumption of u‴(·) =0, which implies that the marginal utility of income is decreasing at a constant rate. This allows us also to adopt a more compact notation. LetΔuY:=u(Y^ik)u(Yik)>0

Cross-border health care

In this section, we analyse how policies intended to increase cross-border patient mobility is likely to affect quality provision and social welfare in each of the three regions. We do so by conducting comparative statics with respect to each of the two mobility cost parameters; the non-monetary cost F and the monetary cost π. A reduction in the patient copayment, π, has a straightforward policy interpretation, while a reduction in F can be interpreted as a policy to reduce the ‘red tape’ costs

Income inequality

In this section we exploit the structural richness of our model to analyse how regional quality provision depends on the degree of income inequality – both across and within regions – when patients have the option to seek treatment outside their own region.

Conclusions

Cross-border patient mobility is an important issue across countries – as exemplified by the new regulation in the EU – and across regions within countries with regional health-care provision, such as Canada, Italy and Sweden. In this paper, we study the consequences of cross-border patient mobility on the quality of health care and the corresponding regional welfare effects. We develop a Salop model with three regions; a high-income, a middle-income, and a low-income region. In each region,

References (23)

  • W. Beckert et al.

    Choice of NHS-funded hospital services in England

    Economic Journal

    (2012)
  • Cited by (24)

    • Cross-border healthcare: A review and applicability to North America during COVID-19

      2022, Health Policy OPEN
      Citation Excerpt :

      Another group of authors utilized mathematical modelling to demonstrate the concerns of cross border healthcare. They argued that cross-border healthcare would benefit the larger and more wealthy countries, and that it would cause the quality of the care to fall in the smaller and poorer countries [47]. Despite these concerns, movement of patients between healthcare systems has been less than anticipated by these countries, which ultimately reduced the overall anticipated negative impacts.

    • Hospitals’ strategic behaviours and patient mobility: Evidence from Italy

      2021, Socio-Economic Planning Sciences
      Citation Excerpt :

      show that if regions simply differ in their skills, cross-border movement may be beneficial for high and low skill regions alike because it provides a strong incentive to stimulate quality improvements so that in the long run inter-regional mobility should disappear [17]. However, when regions differ in income and financing abilities, patient mobility may have adverse effects on welfare [18]; this result is in line with [19]. A number of empirical contributions explain patient flows, most of them focusing on patient choices [7,8,20], but strategic behaviour by hospitals has been almost entirely neglected.

    • Effects of healthcare quality and reimbursement rate in a hospital association

      2021, Socio-Economic Planning Sciences
      Citation Excerpt :

      In the subsequent sections, we discuss where our paper fits in each research stream and present the contribution it makes. Within the patient mobility literature, there is a limited but growing body of work that analyzes the impacts of patient mobility on the operational performance of healthcare/hospital delivery; see the recent papers by Andritsos and Tang [10,19] and Brekke et al. [26,27]. Andritsos and Tang [10,19] use a game-theoretical queueing model to examine how patient mobility affects waiting times and social welfare.

    • Medical travel of cardiovascular and cerebrovascular diseases inpatients in central China

      2021, Applied Geography
      Citation Excerpt :

      The geographic limits associated with and medical facilities covered by government subsidized health insurance programs can shape the spatial pattern of domestic medical travel (Chakrabartia & Tatavarthy, 2019). Our findings suggest that China's new health insurance policy and financing model have increased patients' access to health facilities outside their home environment and stimulated medical travel, which is generally consistent with recent studies (Brekke et al., 2016; Pan & Moreira, 2018). Accessibility/transportation is also an influential factor on CCD patients' mobility.

    • Does fiscal decentralization affect regional disparities in health? Quasi-experimental evidence from Italy

      2019, Regional Science and Urban Economics
      Citation Excerpt :

      Patient-inflow from poor regions would then have increased, exacerbating disparities in the Mezzogiorno among patients able to migrate to other regions (likely the richer) and those who are not (the poorer). Brekke et al. (2016) provide a similar argument in a model where regions are fully responsible for financing their healthcare systems and spending is completely funded with local resources. In particular, the authors show that, if inter-regional inequality increases, then we will observe an increase in the quality of services in high-income regions and an opposite effect in low income regions.

    View all citing articles on Scopus

    This work was carried out with funding from COMPETE reference no. POCI-01-0145-FEDER-006683, with the FCT/MEC's financial support through national funding and by the ERDF through the Operational Programme on “Competitiveness and Internationalization – COMPETE 2020 under the PT2020 Partnership Agreement. The paper has benefited from being presented at the European Health Economics Workshop 2015 in Toulouse. We are grateful for comments by Christoph Schottmüller, Helmuth Cremer, and three anonymous referees.

    View full text