Free for Caring? The Effect of Offering Free Online Medical-Consulting Services on Physician Performance in e-Health Care

Abstract

Introduction:

In online medical-consulting platforms, physicians answer medical questions and share health care knowledge with patients, which will bring them both economic returns and social returns. Little research has studied the free online medical-consulting services (FOMCS) and their influences on the performance of patients.

Methods:

An OLS specification-regression model with robust standard errors clustered at the physician level based on panel data in physician-month level, including 48,406 physicians in 4,452 hospitals in 139 months (August 2006–December 2017) from the largest online health care platform in China.

Results:

We find that FOMCS positively affects the performance of a physician. Specifically, we found that the 1% increase of FOMCS would help physicians receive an additional 3.1% paid consulting services and an 11.4% increase in income. In addition, this effect is enhanced for physicians due to a high pricing strategy and weakened for physicians with high platform-usage experience and professional titles.

Conclusion:

We provide evidence on the effect of FOMCS of physicians on their performance and explore this heterogeneous effect in three kinds of mechanisms: pricing of online medical-consulting services, usage experience of online health care platform, and the professional title of a physician. FOMCS can help quality physicians show their skilled abilities to potential patients to attract their business, make up for the lack of usage experience in online health care platforms, and provide an alternative way to build physicians' online reputations. Our results have implications for existing health management and e-health literature.

Introduction

An online health care platform, as an internet-based virtual platform, has been viewed as a key solution to the shortages of the health care industry, such as medical-resource allocation and physician–patient interaction.¹ It could enhance the relationship between patients and physicians by providing an alternative consultation channel, different care methods, and favorable treatment circumstances.² In online health care platforms, physicians answer medical questions and share health care knowledge with patients in their spare time, which can enable them to meet the medical demands that offline health care providers cannot fulfill and bring them both economic returns and social returns.³ In this regard, for instance, if the patients are located in remote areas from a hospital or might feel an urgent need to visit a hospital personally, the online health care service can provide a more convenient and easy approach for patients to access health services. Online medical-consulting services are novel channels through which physicians can offer more intensive interactions to patients at lower costs. It is not only a place where patients can ask medical problems but also where patients can get references from which to choose appropriate online physicians. Therefore, the online health care service extends traditional offline health services and satisfies the unfulfilled medical demands that offline health care fails to accomplish.

According to information, there is an asymmetry of health care knowledge between physicians and patients.⁴ Online medical-consulting services could promote better interactions that generate feelings of personal gratitude and trust.⁵ Practically, patients usually lack the knowledge for diagnostic skills and treatment methods. Meanwhile, compared with on-site physicians, online medical-consulting services have health care specialists with the same levels of proficiency but lower treatment prices. Given this situation, patients are willing to pay adequate fees to access this service. According to the rules of the online medical-consulting service website, physicians must reply to the paid questions of users as soon as possible. In contrast, for those unpaid users' questions, physicians always answer selectively because they do not have to answer all the free questions. In fact, users can decide whether to pay physicians to achieve guaranteed replies before submitting consultations. Further, physicians can decide whether to answer unpaid consultations, and they must answer those paid ones. At the same time, a portion of online physicians would like to provide free online medical-consulting services (FOMCS), as this kind of online personal reputation can be accumulated and broadly recognized by patients through the internet.⁶ Both of free and paid consultation services are showed in online Q&A format. The FOMCS could benefit physicians in a nonprofit way, such as social returns, thereby further improving the physicians' performance.

IS researchers show great interest in health care information technology and the online health care community (OHC).⁷ However, little research has studied FOMCS and its influences on the medical-treatment activities of patients. The emergence of FOMCS in online health care platforms provides more information for patients to choose online physicians, and, at the same time, it influences physicians' online medical activities.

Thus, an unsolved question is this: Can FOMCS always improve physicians' performance? From an economic point of view, some researchers proposed the value-discounting hypothesis, which argues that by offering free promotions, products/services will be valued less by consumers with lower purchase intentions.⁸ Sometimes, the “free” services may generate potential risk of losing customers when they perceive that the quality of services/products does not satisfy their needs and that the cost of the free services/products is low. As a result, customers may not pay for the services/products, or lower the price they are willing to pay. Moreover, the launching of free service usually increases the financial and time costs of the providers, which will make their performance unpredictable. On the contrary, a number of researchers indicated that free products/services could improve the actual performance of sales by serving as a source of information about the underlying value of the products/services.⁹ Studies show that “free sampling” increases immediate small purchases,¹⁰ and future total payments of consumers¹¹ as the quality of service and the taste of buyers are improved.

Although prior studies offered us some knowledge to understand how such free services will influence actual performance from the domains of marketing, consumer behavior, and management, little research has been carried out in the field of online free services on physicians' performance in health care. An unanswered question is how FOMCS affects physicians' future performance.

Given the conflicts and the limited information on how physicians could improve performance in online health care platforms, this research explores and reveals the actual impacts of FOMCS on physicians' performance. Taking advantage of 1,397,194 observations and more than 54 million online medical-consulting service records on a large Chinese online health care community from August 2006 to December 2017, this study empirically tests the relationship between FOMCS and the performance of physicians. First, FOMCS could improve physicians' performance in general. Second, the effectiveness of FOMCS on performance increases when a physician sets higher medical-consulting service fees. Third, the FOMCS can also help less experienced physicians to earn extra benefits in terms of online healing and physician–patient interaction abilities, as well as those with low professional titles.

This article may contribute to the literature in several ways. First, to our knowledge, this is the first study to show the role of FOMCS in e-health care platforms, which adds to both the streams of the health IT and online communities. Second, FOMCS is different across various countries. Therefore, the online physician–patient activities in China can provide us a unique and representative setting to investigate FOMCS and its impacts in various groups of observations. Third, we highlight the facilitating roles of professional titles, usage experiences of the platform, and pricing strategy by revealing their moderator effects in online physicians' performance.

Literature Review

The Performance of Physicians in Online Medical-Consulting Services

Health care service provides a more effective approach for patients to solve their concerns and communicate with physicians.^12,13 Meanwhile, physicians are able to track the patients' conditions and increase their revenues, which is hard to achieve offline.¹⁴ Online health care service is considered an important supplement for offline hospitals because of the serious problems with the unbalanced numbers of physicians and patients.^15,16 Therefore, much research has arisen to investigate the socioeconomic value and the inhibitors of this emerging information technology, as well as the relationship between patients and physicians.^3,16

Researchers explored the traits and performances of physicians in online medical-consulting services from different perspectives. For example, Liu and Ye⁴ and Chevalier and Mayzlin¹⁷ indicated that the reputation of physicians and hospitals could positively affect physicians' online performance. Guo³ pointed out that physicians could improve their performance by gaining socioeconomic returns from patients in online health care communities. Yang et al.¹⁸ investigated system-generated information of physicians' service quality, which positively influences patients' decision making. Liu and Ye⁴ examined the role of price in online health-consulting purchases and perceived quality based on the exposed information about physicians' abilities. Price is also regarded as an essential signal for product/service quality.¹⁹ In addition, usage experience can enhance the utilization of IT and the performance of users²⁰ as well as the perceptions of users.²¹ The level of a physician's title can influence the perceived service quality of patients and further influence their decision making.⁴

Decision-Making Activities when Offering Free Promotion

Free offering or free sampling is considered an effective promotion method in many cases.^9,22,23 Palmeira and Srivastava²⁴ found that free offerings are more instrumental than low prices and discounts to increase consumers' willingness to pay for the product after the promotion. Mahadevan²² indicated that the fundamental philosophy behind free service was giving up today's revenues in return for assured future revenues. Kolyesnikova and Dodd⁹ further examined that free offerings can improve actual sales performance by serving as a source of information about the underlying value of the products/service. Jiang and Sarkar²⁵ discussed the optimal number of free adopters in software diffusion depending on the discount rate, the length of the demand window, and the ratio of low-valuation to high-valuation free adopters.

On the other hand, Friedman²⁶ indicated that “there's no such thing as a free lunch” in his book. The principle behind this statement is intuitive. Free offerings may place customers in disadvantaged situations.²⁶ Some researchers found that products/services could be valued less by consumers with lower purchase intentions by offering free promotions.⁸ Sometimes, the “free” may generate a potential risk of losing customers when they perceive the quality of services/products to not satisfy their needs and perceive that the costs of the free services/products are low. There is no consistent conclusion on the effects of offering free services.

With the increasing popularity of the online health care community, the platform provides a source of information available for patients to search for suitable physicians and obtain other patients' feedback and recommendations.¹⁶ Offering free service in online health care communities is similar to the mechanism in the traditional markets. However, significant differences exist between them, such as the degree of information asymmetry. Patients may face more serious information asymmetry, which makes the real quality of physicians' services unclear before the consultation.¹⁸ Thus, this research attempts to explore how the performance of physicians will increase when they offer free services.

Hypothesis Development

The Impacts of FOMCS on Physician Performance

The Elaboration Likelihood Model (ELM) provides us a good perspective to study the relationship between FOMCS and physician performance. It describes how consumer attitudes and product engagement could influence their information processing.²⁷ According to the ELM, online patients make their physician-choosing decisions according to the judgment of the physician quality (central) and medical-related (peripheral) cues. Thus, FOMCS provides perfect places for physicians to present these cues to online patients and view their own treatment-quality feedback from online patients. Through this two-way interaction, physicians could enhance their service quality cumulatively and further improve their performance.

Practically, patients have various levels of treatment knowledge, personal goals, and motivations to elaborate on the information presented on the online health care platform. According to the ELM, these variations drive information recipients to collect more cues before decision making, especially when they perceive insufficient cues in central routes in individual tasks.²⁷ Given this condition, FOMCS allows patients to acquire treatment information from target physicians and other patients to assist or view their own physician-choosing behaviors.

The Social Comparison Theory²⁸ explains that it is not always possible for people to get access to objective cues and standards directly, so they tend to compare themselves with others in similar situations due to their internal drive.²⁸ Medical problems proposed in the context of FOMCS by other patients provide valuable information for new patients to realize and understand their own positions to choose suitable online physicians to solve their specific health issues. The appropriate matches and common consensus between physicians and patients could improve the effectiveness of treatment.²⁹ In short, FOMCS could improve physician performance by enhancing the quality of physician treatment and patients' knowledge about the underlying physicians' ability and their matching fitness. Thus, we hypothesize:

H1: The frequency of physician FOMCS participation is positively associated with their medical performance.

The Traits of Physician as Moderators

Self-price-setting of medical treatment

Price is an essential indicator that is always seen as a signal of product/service quality.¹⁹ A good and experienced physician with a higher service pricing usually has fewer patients, as patients may feel potential uncertainty or feel unworthy of the expected medicine service on e-health care platforms. However, FOMCS provided by high-priced physicians could offer the opportunities for potential patients to perceive their medicine quality. Being consistent with Mahadevan²² as well as Palmeira and Srivastava,²⁴ FOMCS increases the willingness of patients to pay for services when they perceive good quality. Meanwhile, when unskilled physicians (with low prices) offer FOMCS, patients may make the inference that the abilities of the physicians are low. This should lower the prices that patients are willing to pay for the treatments. The results of “free” will become less effective or “value-discounting” when the original price of service is low.⁸ Thus, we propose:

H2: The list price strengthens the effect of FOMCS on physician performance.

Online health care–platform usage experience

Social network and network positions influence individual behavior and performance,³⁰ and these effects are also valid in online health care platforms.¹ Physicians would build their own new social networks through the usage of online health care platforms, and these networks will bring physicians more patients. The FOMCS function of online platforms will help new entering physicians to access potential patients via building their own social networks rapidly. With the usage of platforms, networking resources could be obtained by physicians' approaches to saturation. Thus, the influence of FOMCS on physician performance is less than it was earlier. We hypothesize as follows:

H3: The online health care–platform usage experience weakens the effect of FOMCS on physician performance.

Professional title of physician

In most cases, physicians know more about patients' health conditions than patients themselves, according to information asymmetry.³¹ Positive individual reputation is positively associated with consumers' purchase decisions in many online activities.¹⁷ The online services offered by high-title physicians are usually considered “credence services.” Thus, a physician with a high professional title will attract patients with a solid online reputation. Low-title physicians are usually not able to obtain such online trust and reputation as high-title physicians do. In this case, FOMCS provides them an alternative way to attract patients and build their online reputations, as patients could get access to patient–physician interaction quality through the Q&A context in FOMCS. The attention of patients would be drawn when they perceive more FOMCS, and an accumulated individual reputation will increase the number of appointments.⁴ These reputation-building effects will become more efficient in the group of low-title physicians than the high-title group. Thus, we posit:

H4: The level of professional title weakens the effect of FOMCS on physician performance.

Data and Research Model

Data Collection and Measures

This study collects data from a leading online health care platform named Haodf in China. This platform has data of 529,000 physicians from 8,241 hospitals. The functions of the platform include online medical consultation, appointment referral, medical-information inquiry, knowledge sharing of medical science, physician recommendations, and so forth. In addition, this platform offers a unique institutional setting to separate free and paid consultations. The former is to receive slow and limited response from physicians, and the latter is to immediately communicate with physicians. We selected 48,406 physicians in 4,452 hospitals who had provided paid-consultation services as the target sample. We developed a crawler to collect the historical data of physicians in 139 months (August 2006–December 2017) and their attribute data on January 2018 from Haodf.com. The definitions and statistical descriptions of major variables are shown in Table 1.

Table 1.

Variables Definition and Summary Statistics

VARIABLE	DESCRIPTION	OBSERVATIONS	MEAN	SD	MINIMUM	MAXIMUM
Dependent variables
Performance	Logged number of incomes from online paid consulting services, which is scalar multiplication of the number of paid consultations and the price of a physician in a given month	778,649	0.655	1.724	0	12.602
Transactions	Logged number of paid consulting services	778,649	0.210	0.612	0	6.614
Independent variable
FreeService	Logged number of free consulting of a physician	778,649	1.599	1.957	0	10.788
Moderators
Experience	Logged number of days from starting online consulting services to a given month	778,649	7.577	0.601	5.105	8.189
Title	Official clinic title certified by the national agency with uniform standards. Four stages exist for clinic titles: Archiater (4), Associate Archiater (3), Chief Physician (2), Resident Physician (1), and none (0).	777,464	3.072	0.885	1	4
Price	Logged price of paid consulting service of a physician	778,649	3.313	0.854	1.791	6.907
Control variables
Vote	Logged number of votes showing praise given by patients to a physician	778,649	2.813	1.479	0	7.719
Rating	Rating of a physician	778,297	3.857	0.353	2.4	5
HospLevel	Official certification of hospitals. Six stages of hospital levels exist (coding in order): Third-level grade-A (6), Third-level (5), Second-level grade-A (4), Second-level (3), First-level grade-A (2), First-level (1), and none (0).	778,649	5.497	1.151	0	6
Gift	Logged number of virtual gifts that a physician received in a given month	778,649	0.215	0.618	0	6.077
Letter	Logged number of thanks letters that a physician received in a given month	778,649	0.162	0.470	0	5.273

SD, standard deviation.

Dependent Variables and Independent Variables

Physician performance was measured by the actual revenue (Performance) in a given month as the dependent variable. FOMCS (FreeService) is the key independent variable of interest in our estimation. It is the behavior of the free medical-consulting service offered by a physician in an online health care community, measured by the number of FOMCS of a physician in a given month.

Moderators and Control Variables

List price (Price) indicates the price of consulting established by physicians. Online health care–platform usage experience (Exp) is the period between the start time of a physician entered into the platform and a specified date by day. Professional title (Title) refers to the kind of official clinic title certified by the national agency with uniform standards. Several control variables are considered to ensure the model robustness; examples include Vote, Rating, Hospital level, Gift, and Letter.

Estimation Model

We construct a linear regression model to investigate how the free service of a physician affects the revenue, as given next: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} Performanc{e_{it}} = & { \beta _1}FreeServic{e_{it - 1}} + { \beta _2}FreeServic{e_{it - 1}} \times \;Pric{e_i} \;{ \rm{ \;}} \\ & + { \beta _3}FreeServic{e_{it - 1}} \times \;Experienc{e_i}{ \rm{ \;}} \\ & + { \beta _4}FreeServic{e_{it - 1}} \times Titl{e_i} \;{ \rm{ \;}} + \gamma {C_{it - 1}} + { \mu _i} + { \upsilon _t} + { \varepsilon _{it}} \tag{1} \end{align*} \end{document}

As described earlier, \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$Performanc{e_{it}}$$ \end{document} , being the dependent variable, is measured by the consulting revenue of a physician in a given month. \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$FreeServic{e_{it - 1}}$$ \end{document} is the independent variable. \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$Experienc{e_i}$$ \end{document} , \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$Titl{e_i}$$ \end{document} , and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$Pric{e_i}$$ \end{document} are the moderate variables. \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$${C_{it - 1}}$$ \end{document} is a series of control variables including \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$Vot{e_i}$$ \end{document} , \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$Ratin{g_i}$$ \end{document} , \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$Hospital \;Leve{l_i}$$ \end{document} , \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$Gif{t_{it - 1}}$$ \end{document} , and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$Lette{r_{it - 1}}$$ \end{document} . And \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$${ \mu _i}$$ \end{document} is a physician-level fixed effects to account for time-invariant heterogeneity across tourist attractions, \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$${ \upsilon _t}$$ \end{document} is a month fixed effects to control unobserved temporal trends or shocks that apply to the entire market, and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$${ \varepsilon _{it}}$$ \end{document} is the idiosyncratic error term. Thus, we use Equation (1) to fit our panel data in physician-month level. The treatment effect of offering free services can be found by the sign and significance level of \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$${ \beta _1}$$ \end{document} . The moderating effects are shown in the coefficients \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$${ \beta _2}$$ \end{document} , \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$${ \beta _3}$$ \end{document} , and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$${ \beta _4}$$ \end{document} . Further, we adopt the number of paid consulting services of a physician ( \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$Transaction{s_{it}}$$ \end{document} ) to verify the stability of the results.

Results

We present our regression results from Equation (1) in Table 2. Columns 1–2 show the main effect. To check the consistency of the estimated effects, we run an OLS specification with robust standard errors clustered at the physician level as column (1). In this way, we validate the estimation consistency across models. In the main model (both the OLS and FE results), we find that online free service positively affects the performance of a physician (p < 0.01), supporting H1. According to the FE model in Column (2), the specific revenue of physicians increases by 11.4% when they offer an extra 1% of free consulting services. This result indicates that the exposed quality information of physicians due to online free services certainly drives more paid consulting services to physicians.

Table 2.

Estimation Results

	PERFORMANCE			TRANSACTIONS
	(1)	(2)	(3)	(4)	(5)	(6)
FreeService	0.143^*	0.114^*	0.347^*	0.043^*	0.031^*	0.192^*
FreeService	(0.003)	(0.003)	(0.036)	(0.001)	(0.001)	(0.015)
FreeService × price			0.120^*			0.026^*
FreeService × price			(0.004)			(0.002)
FreeService × experience			−0.080^*			—0.031^*
FreeService × experience			(0.005)			(0.002)
FreeService × title			−0.011^*			−0.005^*
FreeService × title			(0.004)			(0.002)
Physician fixed effects	No	Yes	Yes	No	Yes	Yes
Month fixed effects	Yes	Yes	Yes	Yes	Yes	Yes
Control variables	Yes	Yes	Yes	Yes	Yes	Yes
Observations	750567	750567	750567	750567	750567	750567
Adjusted R-squared	0.415	0.296	0.305	0.393	0.284	0.289

Robust standard errors are enclosed in parentheses.

^*p < 0.01.

We explore the mechanisms that explain this effect by introducing three moderators. Column (3) presents these moderation effects, where the effect of FOMCS on physician performance is enhanced (0.120*) for physicians with high-pricing strategies, supporting H2. It indicates that offering online free services can help high-pricing physicians to show their treatment quality and establish reputations. In addition, this promotional effect is weakened (−0.080* and −0.011*) for physicians with high platform-usage experience and professional titles, supporting H3 and H4. It means that physicians of lower usage experience or professional title can benefit more from providing free services, which demonstrates the positive effect of FOMCS on making up for the lack of usage experience or professional reputation. The former indicates that the FOMCS function of online platform will help new physicians to access potential patients via building their own social networks rapidly. The latter means that the reputation-building effects will become more efficient in the group of low-title physicians than the high-title group. The results are consistent with both the consulting revenue and the number of transactions as dependent variables, as shown in column (4–6).

Conclusion

In response to the increasing growth of online medical-consultation platforms, we provide evidence on the heterogeneous effect of FOMCS of physicians on their performance through pricing, experience, and title. Specifically, we found that the 1% increase of FOMCS would help physicians receive an additional 3.1% in paid consulting services and 11.4% in income. Meanwhile, we explore this effect in three kinds of mechanisms: pricing of online medical-consulting service, usage experience of online health care platform, and professional title of a physician.

First, we found that, because price is always seen as a signal of product quality, higher pricing of consulting services could benefit physicians more from FOMCS. It means that increased FOMCS can help quality physicians show their skilled abilities to potential patients and attract their business. Second, we found that the online health care platform–usage experience would weaken the promotional effect of FOMCS on physicians' performance. It indicates that FOMCS can make up for the lack of usage experience in online health care platforms and engage patients to understand the real effectiveness of treatment shown in past FOMCS.

Finally, in regard to professional title, physicians with relatively low levels of professional title benefit more from FOMCS. In this case, FOMCS provides an alternative way to build physicians' online reputations, as patients could get access to quality patient–doctor interactions through the Q&A context in FOMCS.

Implications

This study contributes to the literature in several ways. First, to our knowledge, this is the first study to show the role of FOMCS in e-health care platforms, which adds to both streams of e-health care and online communities. The extant studies primarily examined the adoption intention and actual behavior.^12,13 Our research extends the current understanding of the online service-diffusion process through the consideration of FOMCS offered by physicians.

Second, we enrich the existing literature of free offerings in an online context. Offering free services in an online health care community is similar as in traditional markets.^9,18 This research examined whether and how the counseling performance of a physician will increase when they offer free services, which extends the extant studies of free-offering or free-sampling research. The results confirm the effects of free offerings on online business performance and encourage future studies to consider it as an important perspective when studying e-service performance.

Third, this study deepens the literature of e-service by considering free-offering consulting services in specific mechanisms. We found that free offerings can make up for relatively high self-pricing, low usage experience of online platforms, and low reputation of professional title. This study illuminates that offering free consulting services is proved to be an effective way to present real quality and build reputation by interaction with consumers, which is an important insight in existing literature of both marketing and e-health. This provides novel insights into the future studies that tend to take specific business processes into account when studying online health care.

This study also offers some practical implications. First, this study indicates an effective approach to increase income for physicians by enhancing FOMCS. This can prompt physicians to improve their allocations of service based on limited time and budget. Second, our findings will shed light on the facilitating roles of physician traits by testing several practice-oriented variables (price-setting, professional title, and platform-usage experience), which provide valuable implications to practitioners. Physicians of different types can take corresponding measures to promote themselves by FMOCS in online platforms. Due to the imbalance issue of increasing online medical demands and limited e-health system resources globally, the FOMCS will be an effective way to make up for paid services online and offline.

Footnotes

Acknowledgments

This work was supported by the National Science Foundation of China (71532004, 71490720, 71490724, 71601038) and the National Social Science Foundation of China (14CXW045). Dr. Yiyang Bian is designated as joint first author of this article.

Disclosure Statement

No competing financial interests exist.

References

Venkatesh

, Zhang

, Sykes

. Doctors do too little technology: A longitudinal field study of an electronic healthcare system implementation. Inform Syst Res, 2011; 22:3.

Dedding

, Van Doorn

, Winkler

, Reis

. How will e-health affect patient participation in the clinic? A review of e-health studies and the current evidence for changes in the relationship between medical professionals and patients. Soc Sci Med, 2011; 72:49–53.

Guo

, Guo

, Fang

, Vogel

. How doctors gain social and economic returns in online health-care communities: A professional capital perspective. J Manag Inform Syst, 2017; 34:487–519.

Liu

, Ye

. How price affect online purchase behavior in online healthcare consulting? Perceived quality as a mediator. International Conference for Smart Health. Phoenix, AZ, 2015; 17–18.

Bierstedt

, Blau

. Exchange and power in social life. Am Sociol Rev, 1965; 30:5.

Lim

, Sia

, Lee

, Benbasat

. Do i trust you online, and if so, will i buy? An empirical study of two trust-building strategies. J Manag Inform Syst, 2006; 23:233–266.

Agarwal

, Gao

, DesRoches

, Jha

. Research commentary—The digital transformation of healthcare: Current status and the road ahead. Inform Syst Res, 2010; 21:4.

Raghubir

. Free gift with purchase: Promoting or discounting the brand?. J Consumer Psychol, 2004; 14:1–2.

Kolyesnikova

, Dodd

. There is no such thing as a free wine tasting: The effect of a tasting fee on obligation to buy. J Travel Tourism Marketing, 2009; 26:806–819.

10.

Lammers

. The effect of free samples on immediate consumer purchase. J Consumer Marketing, 1991; 8:2.

11.

Peitz

, Waelbroeck

. Why the music industry may gain from free downloading—The role of sampling. Int J Industrial Org, 2006; 24:5.

12.

Taani

. Patient involvement in medical decisions on telemedicine platforms: The role of social presence and mindfulness. International Conference on Information Systems. Seoul, South Korea, 2017; 10–13.

13.

Yim

, Khuntia

, Lim

, Duan

. To ask or not to ask? An investigation of user engagement and doctor-seeking decision in online health infomediary. Proceedings of the 38th International Conference on Information Systems Seoul, South Korea, 2017; 10–13.

14.

Ziebland

, Chapple

, Dumelow

, Evans

, Prinjha

, Rozmovits

. How the internet affects patients' experience of cancer: A qualitative study. BMJ, 2004; 328:564.

15.

Abouzahra

, Guenter

, Tan

. Integrating information systems and healthcare research to understand physicians' use of health information systems: A literature review. International Conference on Information Systems. Karpacz, Poland, 2015; 20–22.

16.

Zhang

, Gu

, Guo

. The impact of introductory incentives on new users: Evidence from an online health community. Proceedings of the 38th International Conference on Information Systems. Seoul, South Korea, 2017; 10–13.

17.

Chevalier

, Mayzlin

. The effect of word of mouth on sales: Online book reviews. J Marketing Res, 2006; 43:345–354.

18.

Yang

, Guo

, Wu

, Ju

. Exploring the effects of patient-generated and system-generated information on patients' online search, evaluation and decision. Electron Commerce Res Appl, 2015; 14:3.

19.

Wolinsky

. Prices as signals of product quality. Review Econ Stud, 1983; 50:4.

20.

Thompson

, Higgins

, Howell

, Thompson

, Higgins

, Howell

. Influence of experience on personal computer utilization: Testing a conceptual model. J Manag Inform Syst, 1994; 11:1.

21.

Wiedenbeck

, Davis

. The influence of interaction style and experience on user perceptions of software packages. Int J Hum Comput Stud, 1997; 46:5.

22.

Mahadevan

. Business models—Business models for internet-based E-commerce: An anatomy. Calif Manage Rev, 2000; 42:55.

23.

Wang

. How consumers perceive free offers: Implications for mobile marketing. Int J Mobile Marketing, 2007; 2:2.

24.

Palmeira

, Srivastava

. Free offer ≠ cheap product: A selective accessibility account on the valuation of free offers. J Consumer Res, 2013; 40:4.

25.

Jiang

, Sarkar

. Speed matters: The role of free software offer in software diffusion. J Manag Inform Syst, 2009; 26:207–240.

26.

Friedman

. Free offers: A new look. NML Rev, 2008; 38:49.

27.

Petty

, Cacioppo

. Communication and persuasion: Central and peripheral routes to attitude change. New York: Springer-Verlag New York, 1986.

28.

Festinger Leon. A theory of social comparison processes. Hum Relations, 1954; 7:117–140.

29.

Janse

, Gemke

, Uiterwaal

, Van der Tweel

, Kimpen

, Sinnema

. Quality of life: Patients and doctors don't always agree: A meta-analysis. J Clin Epidemiol, 2004; 57:653–661.

30.

Borgatti

, Cross

. A relational view of information seeking and learning in social networks. Manage Sci, 2003; 49:4.

31.

Akerlof

. The Market for “Lemons”: Quality uncertainty and the market mechanism. Q J Econ, 1970; 84:3.