Spatial competition has become the natural environment in which to study several problems in healthcare, ranging from hospital care to drug choices, both from a theoretical and an empirical point of view¹.

However, the market for healthcare is characterized by several important features. Consumers may have very weak incentives to seek out value for money (healthcare is mostly free at the point of use), while their willingness to pay for quality is much higher. However, their effort in seeking the best provider may be reduced by a lack of knowledge about the quality of healthcare. Patients are often infrequent users; key information is withheld by producers on both professional and commercial grounds (Siciliani et al., 2017); and their ability to process such information, as well as the cost of obtaining it, can be limited.

In this respect, several healthcare systems have made quality indicators widely available. For example, in the UK (CITAZIONE), while in Italy the National Healthcare Outcomes Program (Programma Nazionale Esiti, PNE) includes a total of 170 outcome or process indicators and provides specific rankings of all hospitals or local health units (European Observatory on Health Systems and Policies, 2017).

However, choice requires searching for information, and the latter has a cost. With the advent of the internet and the diffusion of data at hospital level, the cost of acquiring information—especially for patients who want to travel outside their catchment area—depends on how the latter is obtained. Reliability and speed of internet connections become important factors in influencing search choices by consumers, because much of the information on the internet is dispersed.

High-speed internet allows patients to access a variety of information sources and aggregate information. Therefore, the general availability of high-speed internet may affect patient decision-making more than any one individual source; in this respect, the diffusion of broadband is certainly an important element.

In this paper, we study the impact that broadband diffusion has on patients’ choice to be treated outside their catchment area. To this end, we use data on hospital admissions in Sicily, the biggest Italian island, whose patients have to travel to the mainland if they want to shop around for quality.

We contribute to two strands of literature. First, we contribute to the literature on the effect of increased information in healthcare markets, where information asymmetries are a concern. The second contribution is methodological: broadband data cannot be obtained at the patient level, as can the other variables used in these models. For this reason, we propose implementing a variation of the hybrid mixed logit model proposed by Czajkowski et al. (2017) to estimate beliefs within a discrete choice model.

To the best of our knowledge, although these models are often used in the context of the transportation literature, this is the first application concerning patients’ choices.