Background
The Pareto principle (also known as the 80/20 rule) states that, in many cases, approximately 80% of the effects result from 20% of the causes. The Pareto principle is very popular in the economic market, as it indicates that a small proportion (eg, 20%) of products in a market often generate a large proportion (eg, 80%) of sales [1]. For example, a relatively small number of titles by established best-selling authors account for a high percentage of book sales, Billboard’s “top 40” hits account for the majority of radio playlists and music sales, and movie rentals are dominated by a few popular “new releases.”
Not surprisingly, the Pareto principle also applies to the health care service market. In the health care service market, a few of the best hospitals or doctors have a much higher market share than do ordinary hospitals or doctors [2]. Market concentration or the Pareto principle may be an advantage in a business context (eg, in a supermarket or bookstore), but it is not good for the health care industry. A major concern for the health care industry is the limited service capability of each hospital or doctor; this is, of course, not a problem in the product sales context. When the market is highly concentrated on a few very good doctors, the efficiency of health service delivery to society as a whole will be negatively affected. That is, a few very good doctors will be extremely busy, while some other unknown doctors will be idle [3]. Therefore, a low concentrated market with balanced supply and demand is the ideal condition for the health care industry. Many efforts have been made to decrease the concentration of the health care market by balancing supply and demand. One example of such efforts is the role of family physicians in western countries, who make up a sizable portion of the primary care workforce [4]. A patient must see a family physician before seeing doctors at higher-level hospitals. Another example is the hierarchical diagnosis and treatment system in China, an important part of China’s medical reform [5]. To encourage patients to see doctors initially at local, lower-level hospitals, China implemented new policies such as charging less at lower-level hospitals and increasing the rates that medical insurers pay to these hospitals.
A recent trend in eHealth is delivering health care services online [6,7]. Among online health services, E-consultation seems to be particularly attractive and is increasing in popularity [8-10]. This new type of online health care consultation can reduce both waiting time and travel expenses. It is also likely to be a valuable option in terms of providing patients with more efficient diagnoses. Most important, patients who have limited or even no medical resources have equal access to medical experts online, leading to better and more efficient use of nationwide medical resources. Online health care consultation will be of great significance in reducing medical costs and in improving the operational efficiency, effectiveness, and equity of medical resources, as well as in enhancing customer satisfaction. Therefore, the use of E-consultation is becoming popular and has been increasing rapidly in recent years [11]. It still has some downsides such as overreliance on it, which can lead to delays in care delivery, patients’ privacy and confidentiality, and technical difficulties involving such technology. Examples of third party E-consultation websites include askthedoctor.com [12], askdoctorfree.com [13], and haodf.com [14]. E-consultation has also been used by some offline medical institutions, such as the Mayo Clinic, to deliver health care services online [15-17]. E-consultation appears to have improved access to specialists, who can be integrated into care processes when timely expert opinions are needed.
The new technologies embedded in E-consultation are not limited to digital communications, computing, and storage but also involve a qualitative transformation in search tools, recommendation tools, and social network technologies [18]. Therefore, E-consultation not only provides convenience and better decision results for the user but also changes the costs of seeking information and patterns of searching for information. Because the choice of a doctor is made based on the information received, the concentration of the online market will be different from the offline market. Two possible consequences of Internet technology on these concentrations have been observed in the E-commerce market, namely the “Long-Tail” effect and the “Superstar” effect. On one hand, the user can find more niche (unpopular) doctors at much lower searching costs than ever before, creating a “Long Tail” in the concentration of demand for doctors. On the other hand, users can more easily find the most high-profile doctors online, creating “Superstar” or “winner-take-all” markets where some very good doctors dominate the market [19]. However, we have no idea whether the E-consultation market will be more of a superstar market or a long-tail market compared with a traditional offline context. What can website designers do to turn the superstar market into a long-tail market?
In this study, we aim to investigate the following research questions:
RQ1: Will E-consultation be more of a long-tail or a superstar market? Or, will the E-consultation market be less concentrated or more concentrated than the offline market?
RQ2: Can information disclosure mechanisms (the doctor’s online reputation and self-representation) help reduce market concentration?
Research Hypotheses
Star Effect Versus Long-Tail Effect
Choosing a doctor on a website is totally different from choosing a doctor at an offline hospital. A significant difference is the information available to the user when making a decision. With the help of information technology such as search engines, recommendation tools, and social networking technologies, the user can easily reach more doctors (especially unknown doctors) at a much lower cost than before. In the traditional offline context, the user’s choice set of doctors is quite small. The user usually chooses a doctor near their home or workplace. However, in the online context, the user can choose any doctor nationwide with just a few clicks of the mouse. This means that the choice set in the context of E-consultation is much larger, and users have more of an opportunity to choose unknown doctors than ever before. Thus, the online market will be less concentrated on a small number of high-profile doctors, creating a long-tail effect.
Another possible consequence of E-consultation is the superstar effect, also known as the Matthew Effect or “the rich get richer.” This is because popular doctors enjoy greater visibility on E-consultation platforms (eg, they are ranked highly by search engines or recommended preferentially by websites). As a consequence, the very good and popular doctors have a greater chance than before of being identified at the national level, which further increases their chance of being chosen by users. Thus, the online market will be more concentrated on a small number of famous doctors, creating a superstar effect.
In summary, both the long-tail and superstar effects may exist in the E-consultation context. We cannot know which effect will be dominant without an empirical study. Therefore, we propose the following two competitive hypotheses:
H1a: The online market is less concentrated than the offline market.
H1b: The online market is more concentrated than the offline market.
Information Asymmetry Theory and Information Disclosure
Health care is a market with high information asymmetry. Information asymmetry models assume that at least one party to a transaction has relevant information while the others do not. In the case of E-consultation, doctors have more information about their own service quality than do the patients. Although doctors know their own service quality, patients have little information on this very important question. This situation of information asymmetry creates an imbalance of power in transactions, which can sometimes cause the transactions to go awry—a type of market failure in a worst-case scenario.
According to signaling theory [20], information asymmetry can be reduced by one party (termed the agent) credibly conveying information about itself to another party (the principal). The recipient of the information can interpret the received signal and adjust their purchases accordingly. For example, in Spence’s job-market signaling model [20], (potential) employees send signals about their abilities to employers by acquiring education credentials. The informational value of the credential comes from the fact that the employer believes the credential is positively correlated with having greater ability and is difficult for less-able employees to obtain. Thus, the credential enables the employer to reliably distinguish low-ability workers from high-ability workers.
In the E-consultation context, doctors send information about their service quality to patients. After receiving this information, patients may change their judgment about doctors’ service quality and further change their choice of doctors. In this study, we focus on two signals that a doctor can send about their service quality on an E-consultation website, specifically, online reputation and online self-representation.
Doctors’ Online Reputations
An online reputation (also known as online word-of-mouth) is built based on feedback from patients. E-consultation websites usually provide a feature known as “rate a doctor.” A patient who has visited the doctor previously can write a review of the doctor in terms of technical competence, interpersonal manner, systems issues, etc. The online reputation system is very popular on E-commerce platforms and has been demonstrated as a reliable mechanism to reduce market information asymmetry. For example, eBay uses a system of customer feedback to publicly rate each member. Amazon [21] has a similar reputation mechanism in place, and merchants develop their reputations across different dimensions [22]. According to a recent study, a doctor’s online reputation, as rated by patients is a good indicator of that doctor’s service quality in the real world [7].
If the E-consultation website does not provide an online reputation feature, the user judges the doctor’s service quality based only on the doctor’s professional standing (eg, director, associate director). Therefore, the user’s consideration set is small because only those doctors with high offline positions will be considered. When the E-consultation website does provide an online reputation feature, users have more clues to evaluate the doctor’s service quality. If the market has many doctors with strong reputations, users will consider those who are highly rated but perhaps have lower offline positions. This means that the consideration set is enlarged. However, if the market is full of doctors with poor reputations, users will not include those poorly rated doctors in the consideration set. This means the consideration set remains at the same size or is even smaller (if doctors with high positions are poorly rated). Thus, having a market with highly rated doctors is very important. If the market has many highly rated doctors, market efficiency will be improved because users have more credible doctors from which to choose (ie, the supply of high-quality doctors is increased). In the same vein, market efficiency will not be improved if the market has few highly rated doctors. Therefore, we propose the following hypothesis:
H2: A market served by many doctors with strong online reputations is less concentrated than a market served by many doctors with poor online reputations.
Doctors’ Online Self-Representation
Self-representation is the activity a doctor commits online for the purpose of sending quality information. There are several ways for doctors to represent themselves on an E-consultation website. For example, a doctor can post articles or provide free consultation services. Such efforts are another type of signal the doctor sends to users. The user can evaluate the doctor’s service quality in terms of the efforts reflected online. For example, doctors who post popular medical science articles demonstrate not only their medical knowledge and skills but also their positive attitudes toward E-consultation as well as their Internet savvy. Meanwhile, the quality and number of free consultation services provided are good indicators of the doctor’s expertise and social responsibility.
Therefore, when an E-consultation website provides self-representation features, the users have more information with which to judge the doctor’s service quality. If the market has many doctors representing and promoting themselves, users will consider these highly represented doctors, who may not be well known offline. This means the consideration set, as well as the supply of high quality doctors, is enlarged. However, if the market is full of doctors with low self-representation, users will not consider these low-effort doctors, and thus the consideration set remains the same. For the same reason, if the market has many highly represented doctors, market efficiency will be improved because users will have more credible doctors from which to choose. Therefore, we propose the following hypothesis:
H3: A market served by many doctors who are highly represented online is less concentrated than a market served by many doctors who are not well represented online.
In summary, we aim to investigate the concentration of the E-consultation market and how to reduce its concentration through information disclosure mechanisms. We hypothesize that the online market is less or more concentrated than the offline market, and the online reputation or self-representation can be used to reduce market concentration.