Abstract
Online health communities (OHCs) are encouraging doctors to engage in prosocial behaviors to promote sustainable development of OHCs and benefit the user base, which may further influence doctor-level performance. However, it is unclear how prosocial behavior and its intensity affect doctor-level performance, such as doctor-patient relationship, online reputation, and online demand. The purpose of this study is to test the effects of prosocial behavior and its intensity on doctor-patient relationship, online reputation, and online demand. We conducted a series of empirical analyses on 103,208 doctor-level samples from a leading OHC in China. Initially, we applied the propensity score matching (PSM) method to obtain a more balanced sample, then, we employed a negative binomial regression model to analyze the sample after PSM, and finally, we performed a robustness test using a zero-inflated negative binomial regression model and changing the time window. Our research findings indicate that prosocial behavior positively affects the doctor-patient relationship, doctors’ online reputation, and online demand. The intensity of prosocial behavior has an inverted U-shaped relationship with the doctor-patient relationship, online reputation, and online demand. The doctor’s clinical title negatively moderates the effect of prosocial behavior on the doctor-patient relationship and online demand. Therefore, OHC operators can draw on our findings to encourage more doctors to engage in prosocial behavior while suggesting they consider the role of the intensity of prosocial behavior and clinical title.
Similar content being viewed by others
Introduction
The burgeoning growth of online health communities (OHCs) has revolutionized how healthcare information and services are disseminated and accessed (Wang et al. 2023b; Zhou et al. 2022a; Huang et al. 2021; Fan et al. 2023). Platforms like Vitals in the United States, Practo in India, and Wedoctor in China, have amassed millions of users, providing them with an invaluable resource for health-related knowledge and consultations (Yin et al. 2022; Qiao et al. 2021). The sustainable development of OHCs cannot be achieved without the active participation of their members (Zhuo and Wang 2024; Liu et al. 2020), particularly healthcare professionals. Doctors’ efforts in prosocial behaviors such as publishing free articles, videos, and providing free consultation services, are pivotal for fostering a positive community ethos and ensuring access to quality healthcare information (Zhang et al. 2020; Yan et al. 2022; Zhang et al. 2019a). Such prosocial behaviors not only enhance the brand image of these OHCs but also significantly benefit the user base by broadening the range of available healthcare knowledge and services (Zhang et al. 2019a; Guan et al. 2018). However, as highlighted by some scholars, the expectation for doctors to engage in these prosocial activities often comes at a considerable personal cost, requiring substantial time, effort, and the potential lost income (Yang and Zhang 2019; Yan et al. 2022). This dichotomy presents a significant challenge, as doctors’ reluctance to engage in these behaviors could negatively impact patient benefits, in turn, hinder the healthy growth of OHCs. In light of this, it becomes imperative to delve deeper into understanding the effect of prosocial behaviors on doctors’ performance. By elucidating the tangible benefits that these activities may yield for doctors, it is possible to foster a more conducive environment for voluntary engagement.
Prosocial behaviors encompass activities undertaken primarily for the benefit of others or society, even when they incur personal costs (Wang et al. 2023b). Examples include volunteering, sharing, and donating. In the context of OHCs, doctors can engage in prosocial behavior by reducing the price of text consultation services to a very low price or free of charge, since this behavior requires a significant amount of time, effort, and loss of economic benefits for the doctors, but mainly for the purpose of making high-quality services available to a larger number of patients at a very low cost. Previous studies in fields such as psychology and sociology have delved deeply into the various benefits of prosocial behavior for individuals providing such acts (Miles and Upenieks 2022; Weinstein and Ryan 2010; Stehr 2023; Rofcanin et al. 2018; Raposa et al. 2016). However, in the fields of information systems and online healthcare, the impact of doctors’ prosocial behaviors on doctor-level performance has not been fully revealed. Although a limited body of OHC-related research has proved the positive impact of prosocial behaviors (e.g., answering one question for free to the patient) on doctors’ economic performance (Yan et al. 2022; Zhang et al. 2019a; Wang et al. 2023a), it remains unclear how prosocial behavior affects the doctor-patient relationship and doctors’ online reputation. The doctor-patient relationship reflects whether there is a positive and harmonious interpersonal relationship between the doctor and the patient (Liu et al. 2020). The doctor’s online reputation reflects their service quality, response speed, and service attitude, usually measured by a comprehensive online rating (Qiao et al. 2021; Yang et al. 2023b). These two aspects of doctor-level performance are pivotal for maintain the healthy and sustainable development of OHCs and for establishing a better image for doctors (Wang et al. 2023b; Liang et al. 2017; Liu et al. 2020; Yin et al. 2022; Huang et al. 2021; Qiao et al. 2021). However, previous literature only explored the antecedents of these two critical aspects of doctor-level performance from the perspectives of doctor and patient participation, online interactional unfairness, and new service mode (Liu et al. 2020; Huang et al. 2021; Zhang et al. 2019b), the influence of prosocial behavior on these critical aspects of doctor-level performance has not been thoroughly explored. By engaging in prosocial behaviors, doctors demonstrate a readiness to prioritize patient welfare and communicate with patients, which patients may perceive as positive signals of the doctors’ work attitude and service quality (Zhang et al. 2019a). Consequently, doctors’ prosocial behaviors have the potential to positively affect patients’ attitudes towards doctors, decision-making processes, and ultimately enhance the doctor-patient relationship and doctors’ reputation. In order to comprehensively examine how prosocial behavior affects doctor-level performance, our primary objective revolves around the following question:
RQ1: How does a doctor’s prosocial behavior affect the doctor-patient relationship, online reputation, and online demand?
In addition, we would like to further explore the effects of the intensity of prosocial behavior, because different intensities of prosocial behavior may also cause recipients to perceive different levels of provider effort and attitude (Yoo et al. 2023), which in turn affects the benefits received by the provider. In our context, the price reduction level may be perceived by the patients as the intensity of prosocial behavior performed by the doctor, because it reflects the potential economic benefits doctors forgo to engage in prosocial behavior (Andrews et al. 2014). Higher price reductions imply that doctors are willing to pay more for prosocial behavior. This measure directly correlates with doctors’ level of concern for patients’ interest and their willingness to provide high-quality service. The answer of this question can effectively guide doctors in developing their own price reduction strategies. However, the findings of prior empirical studies on the relationship between the intensity of prosocial behavior and doctors’ performance are inconsistent in the context of OHCs. The first category of research found that the quantity of contributions to free services has a positive impact on their economic performance (Yan et al. 2022). The second category of research argued that before a certain threshold, the strength of a doctor’s free services has an increasing effect on economic performance, and after a certain threshold, the strength of a doctor’s free services has a decreasing effect on economic performance (Zhang et al. 2019a; Wang et al. 2023a). Due to these various effects, we further examine how the intensity of prosocial behavior plays a role in the doctor-patient relationship, online reputation, and online demand. Therefore, our primary objective revolves around the following two research questions:
RQ2: How does the intensity of a doctor’s prosocial behavior affect the doctor-patient relationship, online reputation, and online demand?
To determine the boundaries of the effect of prosocial behavior, we also study the moderating effect of doctors’ clinical titles. The doctor’s clinical title represents their experience and expertise (Huang et al. 2021); higher clinical titles often correlate with more satisfying consultations for patients (Fan et al. 2023). Given the significant information asymmetry in online healthcare environments, doctors’ clinical titles can be regarded as a signal of medical competence to reduce information asymmetry. Patients often tend to initiate consultations and give evaluations to doctors with high clinical titles (Huang et al. 2021; Qiao et al. 2021). Thus, doctors who have higher clinical titles may also attract more patients and receive more positive evaluations when engaging in prosocial behaviors than doctors with lower clinical titles. Existing literature has examined the moderating effects of the doctor’s clinical title on the relationship between a doctor’s implementation of other new service modes and a doctor’s online reputation, online demand, and offline demand (Qiao et al. 2021; Fan et al. 2023; Huang et al. 2021). However, a few studies explore the moderating effects of clinical title on the relationship between prosocial behavior and doctor-level performance. Therefore, we propose the following research question:
RQ3: How does a doctor’s clinical title moderate the relationship between prosocial behavior and the doctor-patient relationship, online reputation, and online demand?
Drawing on the literature on doctor-patient relationship, online reputation, and online demand in OHCs, and signaling theory, we proposed a conceptual model associated with three groups of hypotheses to address the above research questions. We collected a large objective dataset of doctors who did and did not engage in prosocial behavior from a leading OHC in China. After conducting a series of empirical analyses, we found that prosocial behavior positively affects doctor-patient relationship, doctors’ online reputation, and online demand. The intensity of prosocial behavior has an inverted U-shaped relationship with the doctor-patient relationship, online reputation, and online demand. The doctor’s clinical title negatively moderates the effect of prosocial behavior on the doctor-patient relationship and online demand. However, the moderating effect of the doctor’s clinical title on the relationship between prosocial behavior and online reputation is not significant.
Theoretical foundation and hypothesis development
Signaling theory in OHCs
Drawing upon signaling theory, signal senders can reliably convey information about themselves to signal receivers by sending signals, which reduces information asymmetry (Fan et al. 2023). Effective signals play a crucial role in enabling individuals to evaluate the quality of a product or service, thereby influencing their decisions (Yin et al. 2022). In the context of OHCs, a notable disparity exists in information access between doctors and patients, with doctors having more information related to healthcare services (Yan et al. 2022). According to signaling theory, signals make it easier for receivers to discriminate between high- and low-quality signals accurately (Fan et al. 2023).
In our context, the implementation of prosocial behavior implies that doctors voluntarily sacrifice a part of their economic interests to provide high-quality services for free or at low-cost, and this service also requires doctors to invest a lot of time and effort (Yin et al. 2022; Zhang et al. 2019a). Therefore, this prosocial behavior is a positive service attitude and service quality signal that can help patients eliminate information asymmetry. The intensity of prosocial behavior can also be interpreted by the patient as a positive signal of the doctor’s willingness to put patients’ interests first and communicate with patients (Andrews et al. 2014).
The effects of doctors’ prosocial behavior in OHCs
According to signaling theory, when a signal is perceived as a positive signal of a hidden type (i.e., service quality), the receiver perceives the signal as positively correlated with the signaler’s ability (Khurana et al. 2019). In our context, doctors’ prosocial behavior refers to doctors’ making significant price reductions in fee-for-service so that patients can access high-quality healthcare services at a very low cost. It requires doctors to invest more time and effort to provide high-quality consultation services, and patients can obtain more health information from the complete consultation process (Fan et al. 2023). Therefore, this behavior conveys a positive signal to patients that doctors are willing to put the patients’ interest first and provide high-quality services, and such signaling can alleviate information asymmetry (Wang et al. 2023a). Online demand refers to the number of online consultations a doctor receives in the OHC (Huang et al. 2021; Qiao et al. 2021; Yin et al. 2022). To access high-quality services, patients will be more inclined to consult doctors who engage in prosocial behaviors (Yin et al. 2022), and thus, these doctors may obtain more online demand. Furthermore, prior research suggests that patients trust their doctors more and develop high-quality online relationships when they feel more helpful and supportive in their online interactions with them (Zhang et al. 2019b). In our context, doctors who engage in prosocial behavior may be perceived by patients as a signal that the doctor is willing to engage in positive doctor-patient communication and make efforts to help solve the patient’s health problems (Wang et al. 2023a). As a result, these doctors may achieve a more harmonious and trusting doctor-patient relationship. Finally, engaging in prosocial behavior may positively impact a doctor’s online reputation. Prior research highlights that high-quality services contribute to a positive brand image and reputation in online contexts (Lin et al. 2022), including the healthcare context (Yin et al. 2022; Huang et al. 2021). By providing high-quality services at low prices, doctors signal their ability to offer excellent service, leaving a positive impression on patients and further enhancing their reputation. Drawing from the preceding discussion, we advance the subsequent hypotheses:
H1a. Doctors’ prosocial behavior positively affects the doctor-patient relationship.
H1b. Doctors’ prosocial behavior positively affects their online reputation.
H1c. Doctors’ prosocial behavior positively affects their online demand.
The effects of the intensity of prosocial behavior in OHCs
In our context, the intensity of prosocial behavior implies the extent of their price reduction. Given that doctors can freely choose the extent of their price reduction, doctors who have lowered their pricing from higher to lower may be considered to have engaged in high-intensity prosocial behavior. Similar to the rationale in the previous section, the overall effect of the intensity of prosocial behavior should also promote the doctor-patient relationship, online reputation, and online demand. However, prior research suggests that the impact of the intensity of prosocial behavior on patients’ choices is unlikely to maintain a linear pattern. For example, Wang et al. (2023a) found that when doctors post too many free articles on the OHC, the number of patients choosing them will decrease because some patients may be skeptical about the motivation for doctors’ prosocial behavior and whether they will spend a lot of time and effort on primary healthcare services. A similar problem exists when the intensity of prosocial behavior is too high in our context. For example, a doctor who lowers prices too much may make patients doubt whether the doctor can provide high-quality healthcare services because the price can also signal the quality of healthcare services in online healthcare (Fan et al. 2023). In addition, according to the social exchange theory, an individual weighs both the benefits and the costs before implementing a behavior (Yan et al. 2016), therefore, excessive price reductions can also lead patients to doubt the rationality of this behavior. In this case, patients may be less likely to choose those doctors with excessive price reductions, less likely to develop an intensive relationship with them, and less likely to rate them highly. Therefore, we propose the following hypotheses:
H2a. The intensity of prosocial behavior has an inverted U-shaped relationship with the doctor-patient relationship.
H2b. The intensity of prosocial behavior has an inverted U-shaped relationship with doctors’ online reputation.
H2c. The intensity of prosocial behavior has an inverted U-shaped relationship with doctors’ online demand.
Moderating effects of doctors’ clinical title
The doctor’s clinical title represents the level of a doctor’s medical competence, helping to reduce information asymmetry between doctors and patients (Qiao et al. 2021). Patients tend to be more inclined to choose experienced and medically competent doctors for treatment (Huang et al. 2021; Guo et al. 2017). The impact of doctors’ implementation of prosocial behavior may also vary by clinical title. Prior research has shown that doctors with lower (vs. higher) clinical titles typically attract fewer patients through OHCs (Huang et al. 2021). At the same time, due to information asymmetry in OHCs, even when doctors with lower (vs. higher) clinical titles provide extra efforts to serve patients, they may still struggle to succeed in attracting patients and increasing visibility (Khurana et al. 2019; Huang et al. 2021; Zhang et al. 2019a; Fan et al. 2023). Thus, even if doctors with lower (vs. higher) titles engage in prosocial behaviors, they may have more difficulty obtaining harmonious doctor-patient relationships, enhancing their online reputation, and attracting patients. Therefore, we make the following three hypotheses:
H3a. Doctors’ clinical title positively moderates the impact of prosocial behavior on the doctor-patient relationship; the higher the clinical title, the more the impact.
H3b.Doctors’ clinical title positively moderates the impact of prosocial behavior on the doctors’ online reputation; the higher the clinical title, the more the impact.
H3c.Doctors’ clinical title positively moderates the impact of prosocial behavior on the doctors’ online demand; the higher the clinical title, the more the impact.
Our hypothesized relationships are displayed in the conceptual model in Fig. 1.
The figure presents all the hypothesized associations in the proposed model.
Method
Research context and data collection
The setting for our research is a prominent OHC in China. By leveraging this OHC, doctors can efficiently allocate their time to deliver healthcare services, sharing healthcare knowledge, reap financial benefits, and enhance their reputation (Guo et al. 2017). At present, this OHC connects 270,000 doctors at 7,800 hospitals in 31 provinces across China and has over 220 million users. As shown in Figs. 2, 3, the OHC has launched a new webpage to present a list of doctors who have engaged in prosocial behavior called “Today’s Free or Low-Cost Consultation”. Doctors are free to decide whether to lower the price of fee-for-service and the extent of the reduction. We chose this OHC for data collection mainly because of the well-established reputation of this OHC in the industry and the comprehensive and detailed doctor-related information it provides, such as clinical title, academic title, the number of online consultations and reviews (Qiao et al. 2021). More critically, the OHC has a strict commenting mechanism. Before patients can leave comments, they must undergo authentication via their cell phone numbers. Moreover, only those who have genuinely undergone an online consultation with a doctor are permitted to post comments about them, thereby enhancing the validity of our research data.
The figure shows the layout of the online health community homepage as well as the location of “Today’s Free or Low-Cost Consultation” module.
The figure illustrates the homepage of a doctor on the online health community, highlighting key elements relevant to our variables.
From April 1, 2023, to July 8, 2023, we collected doctor-level data on doctors who did and did not engage in prosocial behaviors every two weeks. Finally, we obtained 7 periods of doctor-level data for 480 doctors who engaged in prosocial behavior and 14,264 doctors who did not, for a total of 103,208 observations, and the number of observations for the intensity of the prosocial behavior variable was 3349.
Variable descriptions
The dependent variables are the doctor-patient relationship, online reputation, and online demand. First, given that the doctor-patient relationship reflects whether there is a positive and harmonious interpersonal relationship between the doctor and the patient (Zhang et al. 2019b; Liu et al. 2020), we use the number of satisfied reviews that doctor i received in time t to represent the doctor-patient relationship (Relationshipit). After each online consultation, patients will evaluate their satisfaction with the doctor’s service, and the higher the number of satisfaction reviews a doctor receives means the more harmonious the relationship between him/her and the patient. Second, online reputation (OnlineReputationit) is measured by the doctor i’s online ratings in time t (Qiao et al. 2021). This is a comprehensive indicator automatically calculated by the website and is a comprehensive evaluation of the doctor’s service quality. It takes into account patient ratings, patient likes, patient virtual gifts, speed of service and other indicators. The higher the doctor’s online reputation, the higher the recognition he or she receives from patients. Third, online demand (OnlineDemandit) refers to the number of online consultations that the doctor i receives in time t (Huang et al. 2021).
The independent variables are whether the doctor implements prosocial behavior and the intensity of prosocial behavior. Prosocial behavior (ProsocialBehaviori) is a binary variable equal to 1 if doctor i reduces the price of fee-for-service in time t, and 0 otherwise. To measure the intensity of prosocial behavior (Intensityi), we measure the intensity of prosocial behavior in terms of the magnitude of the price reduction. The greater the magnitude of the price reduction, the greater the intensity of doctors’ prosocial behavior.
The moderator variable is the doctor’s clinical title. We chose the doctor’s clinical title as the moderator variable because, in China, a doctor’s clinical title tends to more accurately reflect his or her medical experience and professional status in practice than other titles (Tong et al. 2022; Fan et al. 2023). The acquisition of a clinical title is directly related to a doctor’s actual work performance and professional skills, whereas other professional titles (e.g., academic titles), while also important, tend to reflect other aspects of achievement (e.g., research and teaching accomplishments) (Zhou et al. 2022b). Following Qiao et al. (2021), we categorized clinical titles into two levels: higher (i.e., chief doctor) and lower (i.e., associate chief doctor, attending doctor, resident doctor) clinical titles. We constructed the dummy variable ClinicalTitlei, with ClinicalTitlei = 1 if the doctor has a higher clinical title and 0 otherwise. It is worth noting that, to avoid the loss of valuable information due to too many binary variables, in the main effect analysis, the clinical title was measured as a control variable in the following ways (Li et al. 2021; Fang et al. 2022): 4 = chief doctor, 3 = associate chief doctor, 2 = attending doctor, 1= resident doctor, 0 = otherwise.
Control variables included clinical titles (ClinicalTitlei), academic titles (AcademicTitlei), total number of reviews (ReviewsVolumeit), total number of consultations (OnlineVolumeit), total number of followers (Followerit), total number of gifts (Giftit), and text consultation price (Pricei). AcademicTitlei is measured in the following way: 4 = professor, 3 = associate professor, 2 = lecturer, 1 = assistant, 0 = otherwise. ReviewsVolumeit represents the cumulative total number of reviews by doctor i before time t. OnlineVolumeit represents the cumulative total number of online consultations by doctor i before time t. Followerit represents the cumulative total number of followers by doctor i before time t. If a patient is interested in a doctor and wants to follow him/her for a long time (e.g. to get his/her long term treatment and to obtain the latest articles), he can click on the “Follow” button on the doctor’s homepage and become his/her follower. Giftit represents the cumulative total number of gifts by doctor i before time t. In the OHCs, patients can express their gratitude and recognition to doctors by giving virtual gifts after receiving doctors’ online consultation services. Pricei represents the text consultation price of doctor i at time t. Table 1 shows our description of the variables.
Data analysis method
We analyzed the data using a variety of methods to test our hypotheses. First, we recognized doctors’ decisions to engage in prosocial behavior are not random and may lead to self-selection bias. Therefore, before analyzing the data, we employed the PSM method to enhance the comparability between the groups of doctors who did and did not engage in prosocial behavior. Second, we used negative binomial regression models to test our hypotheses because the dependent variables were overdispersed, and their variances were substantially larger than their means. Additionally, considering the excessive number of zeros in the dependent variable, we utilized zero-inflated negative binomial (ZINB) models to ensure the robustness of our results. Finally, to verify whether our results remain consistent across datasets of different durations, we extended the time window from 2 weeks to 4 weeks to further test the robustness of our results.
Moreover, our data analyses are confirmatory. Initially, we conducted a thorough literature review, identified the research questions and theoretical foundation, and proposed our hypotheses. Subsequently, we collected and processed the data to test these hypotheses. Therefore, the sequence of proposing hypotheses followed by data collection and analysis to verify them emphasizes the confirmatory nature of our study.
In addition, our data analysis also has certain limitations. Specifically, our research design does not constitute a natural experiment because OHC did not launch a new feature of online consultations in our sample period. Consequently, causal claims are not possible with the observational data set. Therefore, the potential causal question is, “Do doctors who offer prosocial behavior have more positive reviews because they offer it, or have they first received satisfied reviews and then started to offer it?” We will also discuss this in research limitations.
Propensity score matching
Given that doctors can decide on their own whether to engage in prosocial behavior, this decision is not random and may lead to issues of self-selection bias (Yin et al. 2022). Those doctors with a strong sense of social responsibility or those with higher expertise and more experience may be more inclined to engage in prosocial behavior. To mitigate this bias, we use PSM (Fan et al. 2023), matching doctors who engage in prosocial behavior with those who do not. This method enables us to control doctor level self-selection by constructing a matched control group, where the doctors in the control group are very similar to the doctors in the treatment group in multiple observable dimensions, but do not engage in prosocial behavior (Liu et al. 2021). Therefore, this approach allows us to fairly compare the two groups engaging in prosocial behavior or not. We initially randomize the sequence of all doctors to ensure that the ordering does not influence the subsequent matching process. Next, we employ logit regression to calculate the propensity scores, with ProsocialBehaviori as the binary outcome variable and a set of observed features as covariates. These features include ClinicalTitlei, AcademicTitlei, ReviewsVolumeit, OnlineVolumeit, Followerit, and Pricei. Based on the propensity scores for each doctor, we employ one-to-four nearest-neighbor matching to match doctors from the treatment group with their counterparts from the control group.
To evaluate the efficacy of PSM, we conduct a pre and post-matching examination of the covariate balance between the treatment and control group doctors. The results in Table 2 show that after matching, the percentage bias between control and treatment groups reduces substantially, and t tests indicate no significant differences in the means of covariates between the treatment and control groups. After PSM, we obtained 53,277 doctor observations, including 3349 treatment group samples and 49,928 control group samples. Table 3 also reports descriptive statistics for the variables. The results of our correlation analysis indicate significant correlations between the independent and dependent variables, and low correlations between the independent and control variables. We assess the multicollinearity of the model using the variance inflation factor (VIF) method, and the results showed that the VIF values were still below the conventional threshold (VIF < 3), which proves that our model does not suffer from serious multicollinearity problems (Li et al. 2021; Fan et al. 2023). All subsequent analyses are conducted based on the PSM samples.
Empirical results
Impact of prosocial behavior and its intensity on doctor-patient relationship, online reputation, and online demand
Two dependent variables—Relationshipit and OnlineDemandit—are overdispersed count variables, and their variances of the count data (483.15, 774.39) exceeded the means (1.83, 2.96), respectively. Although the data in the OnlineReputationit variable contains decimals, the proportion of non-negative integers is close to 90%, and the variance of the count data (22.70) exceeded the mean (3.59). Given that the negative binomial regression model has the advantage of processing overdispersed count data and has been widely used to analyze such data (März et al. 2017; Yang et al. 2023a), we employed the negative binomial regression model in our study. Despite collecting data for 14 weeks, we did not employ the fixed effects model. The rationale behind this decision is that our primary independent variable, ProsocialBehaviori, remains constant throughout the sample period. If a fixed effects model were applied, all time-invariant independent variables would cancel out during the transformation (Fan et al. 2023). We evaluated the impact of prosocial behavior and its intensity on doctor-patient relationship, online reputation, and online demand.
Table 4 presents the robust standard errors at the doctor level for clustering. The results indicate that doctors’ prosocial behavior positively affects the doctor-patient relationship, online reputation, and online demand (β = 1.68, 0.55, and 1.75 with all p’s < 0.001). These findings support H1a, H1b, and H1c, suggesting that doctors who engage in prosocial behavior enhance doctor-patient relationship, doctors’ online reputation, and online demand, respectively. We then check the impact of doctors’ prosocial behavior intensity. We followed the guidelines of previous literature on testing U-shaped or inverted U-shaped relationships (Haans et al. 2016; Karhade and Dong 2021). Initially, the coefficients of Intensity in Model 2, 5, and 8 were positive and significant (β = 0.04, 0.04, and 0.10 with all p’s < 0.05), while the coefficients of the Intensity squared were negative and significant (β = −0.0002, −0.0002, and −0.0014, with all p’s < 0.05). These results fulfill the first criterion for testing an inverted U-shaped relationship. We then performed the U test to assess whether the sides of the slope were steep before and after the turning point. This test formalized that on both sides, the positive and negative slopes were statistically significant (t = 2.37, 2.86, and 4.43 with all p’s < 0.01), satisfying the second criterion for testing an inverted U-shaped relationship. Finally, we found that the turning point occurs at 89.2, 106.3, and 68.5, respectively, within the range of values of Intensity variable (5 ~ 195), which meets the third criterion for validating an inverted U-shaped relationship (see Figs. 4–6). Thus, all these results support H2a, H2b, and H2c, suggesting that the intensity of prosocial behavior has an inverted U-shaped relationship with doctor-patient relationship, online reputation, and online demand (Haans et al. 2016; Karhade and Dong 2021).
The figure presents the inverted U-shaped relationship between the intensity of prosocial behavior and the doctor-patient relationship.
The figure presents the inverted U-shaped relationship between the intensity of prosocial behavior and online reputation.
The figure presents the inverted U-shaped relationship between the intensity of prosocial behavior and online demand.
Furthermore, we examine the moderating effect of clinical title on the relationship between doctors’ prosocial behavior and doctor-patient relationship (H3a), online reputation (H3b), and online demand (H3c). We set the doctors without senior titles as the baseline group, and the results are presented in Table 4. The interaction coefficient of ProsocialBehavior and ClinicalTitle in Model 3 is negative and significant (β = −0.97 with p < 0.001), indicating that the doctor’s clinical title negatively moderates the relationship between prosocial behavior and the doctor-patient relationship. This result supports H3a. The interaction coefficient of ProsocialBehavior and ClinicalTitle is not significant in Model 6, indicating that there is no significant difference in the positive impact of engaging in prosocial behaviors on online ratings between doctors with higher and lower clinical titles. These results do not support H3b. The interaction coefficient of ProsocialBehavior and ClinicalTitle in Model 9 is negative and significant (β = −0.83 with p < 0.001), indicating that the doctor’s clinical title negatively moderates the relationship between prosocial behavior and online demand. This result supports H3c.
Robustness checks
We use two approaches to verify the robustness of the results. Firstly, since the data for the dependent variables is over-dispersed and has too many zeros, the ZINB model is good at handling this type of data (Zhou et al. 2022a). Therefore, we consider using the ZINB model to check the robustness. The specific robustness test results are presented in Table A1 and Table A2 of Appendix A. The coefficients of ProsocialBehavior are positive and significant in Model 1, 4, and 7 of Table A1. The coefficients of Intensity and Intensity squared are positive and negative in Model 2, 5, and 8 of Table A1. The coefficients of ProsocialBehavior×ClinicalTitle are negative and significant in Models 3, 6, and 9 of Table A1. These results are generally consistent with our baseline model and demonstrate the robustness of our results. Secondly, following the approach of Qiao et al. (2021), we change the time window from 2 weeks to 4 weeks, using data from April 29, 2023, to July 1, 2023, for the analysis. Our results are displayed in Table A2, where the results are consistent with the baseline model and appear to be robust.
Discussion and implications
Key findings
The implementation of prosocial behaviors by doctors is a new phenomenon in OHCs, and it is still unclear how prosocial behavior and its intensity affect doctor-level performance. This topic may help OHC operators promote this service mode and attract more doctors to participate in prosocial behaviors in the OHC, contributing to the sustainable development of the OHC. Our findings reveal three critical insights.
Firstly, we find that doctors who implement prosocial behavior enhance the doctor-patient relationship, online reputation, and online demand, which indicates that these doctors experience increased satisfied reviews, online ratings, and online consultations. This result can be explained by signaling theory: By implementing prosocial behavior, doctors may release a positive signal to healthcare consumers that they are willing to invest a lot of time and energy to provide high-quality services at low prices, which largely alleviates the information asymmetry between doctors and healthcare consumers (Fan et al. 2023; Yin et al. 2022). Healthcare consumers are more likely to choose doctors who provide high-quality services (Yin et al. 2022), interact positively with doctors who are willing to put the healthcare consumer’s interests first and thus form favorable relationships and impressions (Wang et al. 2023a; Zhang et al. 2019b). Therefore, the implementation of prosocial behavior may influence healthcare consumers’ decision-making, attracting more healthcare consumers to choose these doctors, establishing a good doctor-patient relationship, and receiving high ratings.
Secondly, our study reveals an interesting finding that the intensity of prosocial behavior has an inverted U-shaped relationship with the doctor-patient relationship, online reputation, and online demand. This finding may be due to the fact that if a doctor makes a significant reduction in price, it can lead healthcare consumers to doubt whether the quality of the doctor’s service is high enough (Wang et al. 2023a), which in turn affects the doctor-patient relationship, online reputation, and online demand, as the price can also serve as an important signal of the quality of healthcare services in the context of online healthcare (Fan et al. 2023). Although these questions are not applicable to all doctors with large price reductions, healthcare consumers are unable to validate their suspicions due to limitations in information accessibility and information processing ability (Wang et al., 2023a).
Finally, contrary to our H3a and H3c, we find that doctors’ clinical titles negatively moderate the effect of prosocial behavior on the doctor-patient relationship and online demand, which seems to be inconsistent with the previous studies (Huang et al. 2021; Guo et al. 2017; Zhang et al. 2019a; Fan et al. 2023). One possible explanation is that doctors’ time and energy on the OHC are limited (Yin et al. 2022), and those doctors with higher titles need to deal with more offline matters and online fee-for-service, and their time and energy invested in free or low-cost consultation is limited. However, those doctors with lower titles need to deal with fewer offline matters and online fee-for-service, and their time and energy invested in prosocial behaviors is greater (Qiao et al. 2021; Huang et al. 2021), so the free or low-cost consultation services they provide may promote better doctor-patient relationships and attract more online demand. Additionally, we find no significant moderating effect of the doctor’s clinical title on the relationship between prosocial behavior and online reputation, which is consistent with the study of Qiao et al. (2021).
Theoretical contributions
This study contributes to the existing literature on prosocial behavior, doctors-patient relationship, online reputation, and online demand in OHCs. Initially, combined with signaling theory, we introduce novel theoretical insights into the effectiveness of prosocial behavior and the intensity of prosocial behavior in OHCs. The implementation of prosocial behavior can serve as a positive signal reflecting a doctor’s willingness to provide high-quality services at a low price and to put healthcare consumers’ interests first. These findings provide further empirical support for the signaling role of prosocial behavior in enabling healthcare consumers to differentiate between high and low-service quality doctors. We also find that the intensity of prosocial behavior had an inverted U-shaped relationship with the doctor-patient relationship, online reputation, and online demand. Accordingly, this study provides empirical support for understanding the role of prosocial behavior and its intensity in OHCs through signaling theory.
Subsequently, despite extensive research on the antecedents of doctor-patient relationships, online reputation, and online demand (Liu et al. 2020; Huang et al. 2021; Fang et al. 2022), the literature has seldom concentrated on the effects of prosocial behavior and its intensity. Our study bridges this gap, providing a novel lens through which to examine the factors influencing doctor-patient relationships, online reputation, and online demand in OHCs. Consequently, our work significantly advances the discourse on how doctors can optimize these doctor-level performance.
Finally, we demonstrate that doctors’ clinical titles can decrease the signaling effects of prosocial behavior on the doctor-patient relationship and online demand. While existing studies have explored the moderating role of clinical titles on the relationship between new service modes and doctor-level performance (Qiao et al. 2021; Fan et al. 2023; Huang et al. 2021), our research pioneers in identifying the boundary conditions for the effects of prosocial behavior on doctor-level performance. Our findings suggest that engaging in prosocial behavior by doctors with low clinical titles promotes better doctor-patient relationships and attracts more online consultations. This insight expands our comprehension of the boundary conditions for the signaling effect of prosocial behavior in OHC settings.
Practical implications
Our findings provide managerial implications for the OHC operators and doctors. Firstly, our findings suggest that engaging in the prosocial behavior positively affects the doctor-patient relationship, online reputation, and online demand. Thus, OHCs interested in enhancing healthcare consumer well-being and improving the doctor-patient relationship but do not yet offer a prosocial service mode could offer this new service mode to doctors and encourage them to actively adopt it. Our findings could also encourage more doctors with the intention of helping more healthcare consumers access healthcare services to join OHCs.
Secondly, our study shows that the intensity of prosocial behavior has an inverted U-shaped relationship with the doctor-level performance. Given that price is also an important signal reflecting the quality of doctors’ healthcare services, doctors should not blindly pursue significant price reductions, but should carefully consider the magnitude of price reductions when making price reductions of fee-for-service (Fan et al. 2023). It is important to note that price reductions that are too high may cause healthcare consumers to doubt the quality of doctors’ healthcare services.
Thirdly, our study emphasizes the moderating role of doctors’ clinical titles. Our findings suggest that doctors with lower clinical titles who engage in prosocial behavior obtain better doctor-patient relationships and online demand. In other words, healthcare consumers are more likely to consider doctors’ efforts in OHCs, especially for doctors with lower clinical titles. Our findings provide guidance on how doctors can choose from various service modes in the OHC when considering their medical competence.
Finally, our results can also be generalized to OHCs in other countries when the following essential conditions are met simultaneously: (1) Both doctors and patients extensively use and trust OHCs, which ensures that the OHC system is well-received and has a sufficient user base. (2) Doctors are free to choose whether to engage in such prosocial activities without constraining the pricing of medical services, thereby limiting the space for doctors to provide free or low-cost services. (3) The recognition and appreciation of professional titles in the culture and social structure of the patient’s country is similar to the background of our study, which ensures that the dynamics observed between prosocial behaviors and doctor-level performance are likely to be replicated. Countries that meet these conditions are more likely to extend our research findings to their OHC environment successfully.
Limitations and future directions
This study contains several limitations that may provide fruitful paths for research. Firstly, since we found only one OHC on which doctors engaged in prosocial behavior by offering free or low-cost consultations, we only collected data on this OHC. Future research should investigate whether these findings can be generalized to other OHCs. Secondly, due to the limitation of the OHC’s data, we could not obtain more data related to healthcare consumer characteristics, and future studies could consider collecting consumer-related data from sources such as the hospital’s official website in order to incorporate healthcare consumer characteristics into the research model. Moreover, we thought it would be interesting to explore the impact of changes in prosocial behavior on doctor-level performance, however, since the data collection period was only 14 weeks, during which doctors were not observed to engage or drop out of prosocial behaviors, we could collect data for a longer period of time in the future to explore the impact of changes in prosocial behavior. Finally, a potential causal question is, “Do doctors who offer prosocial behavior have more positive reviews because they offer it, or have they first received satisfied reviews and then started to offer it?” Future research could employ a quasi-natural experiment research design to understand the dynamic relationship between doctors’ prosocial behaviors and patients’ satisfied reviews, and how this relationship evolves over time.
Data availability
The datasets analyzed during the current study are not publicly available because they constitute an excerpt of research in progress but are available for authors’ reasonable request.
References
Andrews M, Luo X, Fang Z et al. (2014) Cause marketing effectiveness and the moderating role of price discounts. J Mark 78(6):120–142. https://doi.org/10.1509/jm.14.0003
Fan W, Zhou Q, Qiu L et al. (2023) Should doctors open online consultation services? An empirical investigation of their impact on offline appointments. Inf Syst Res 34(2):629–651. https://doi.org/10.1287/isre.2022.1145
Fang J, Wen L, Ren H et al. (2022) The effect of technical and functional quality on online physician selection: Moderation effect of competition intensity. Inf Process Manag 59(4):102969. https://doi.org/10.1016/j.ipm.2022.102969
Guan T, Wang L, Jin J et al. (2018) Knowledge contribution behavior in online Q&A communities: An empirical investigation. Comput Hum Behav 81:137–147. https://doi.org/10.1016/j.chb.2017.12.023
Guo S, Guo X, Fang Y et al. (2017) How doctors gain social and economic returns in online health-care communities: a professional capital perspective. J Manag Inf Syst 34(2):487–519. https://doi.org/10.1080/07421222.2017.1334480
Haans RF, Pieters C, He ZL (2016) Thinking about U: Theorizing and testing U‐and inverted U‐shaped relationships in strategy research. Strategic Manag J 37(7):1177–1195. https://doi.org/10.1002/smj.2399
Huang N, Yan Z, Yin H (2021) Effects of online–offline service integration on e-healthcare providers: A quasi-natural experiment. Prod Oper Manag 30(8):2359–2378. https://doi.org/10.1111/poms.13381
Karhade P, Dong JQ (2021) Information technology investment and commercialized innovation performance: Dynamic adjustment costs and curvilinear impacts. MIS Quart 45(3):1007–1024. https://doi.org/10.25300/MISQ/2021/14368
Khurana S, Qiu L, Kumar S (2019) When a doctor knows, it shows: An empirical analysis of doctors’ responses in a Q&A forum of an online healthcare portal. Inf Syst Res 30(3):872–891. https://doi.org/10.1287/isre.2019.0836
Li CR, Zhang E, Han JT (2021) Adoption of online follow-up service by patients: An empirical study based on the elaboration likelihood model. Comput Hum Behav 114:106581. https://doi.org/10.1016/j.chb.2020.106581
Liang C, Gu D, Tao F et al. (2017) Influence of mechanism of patient-accessible hospital information system implementation on doctor–patient relationships: A service fairness perspective. Inf Manag -Amster 54(1):57–72. https://doi.org/10.1016/j.im.2016.03.010
Lin S-W, Huang EY, Cheng K-T (2022) Understanding organizational reputation formation in mobile commerce. Electron Commer R A 55:101200. https://doi.org/10.1016/j.elerap.2022.101200
Liu QB, Liu X, Guo X (2020) The effects of participating in a physician-driven online health community in managing chronic disease: Evidence from two natural experiments. MIS Quart 44(1):391–419. https://doi.org/10.25300/MISQ/2020/15102
Liu X, Zhou Y-W, Shen Y et al. (2021) Zooming in the impacts of merchants’ participation in transformation from online flash sale to mixed sale e-commerce platform. Inf Manag -Amster 58(2):103409. https://doi.org/10.1016/j.im.2020.103409
März A, Schubach S, Schumann JH (2017) Why would I read a mobile review?” Device compatibility perceptions and effects on perceived helpfulness. Psychol Mark 34(2):119–137. https://doi.org/10.1002/mar.20979
Miles A, Upenieks L (2022) Moral self-appraisals explain emotional rewards of prosocial behavior. J Happiness Stud 23(5):1793–1814. https://doi.org/10.1007/s10902-021-00434-w
Qiao W, Yan Z, Wang X (2021) Join or not: The impact of physicians’ group joining behavior on their online demand and reputation in online health communities. Inf Process Manag 58(5):102634. https://doi.org/10.1016/j.ipm.2021.102634
Raposa EB, Laws HB, Ansell EB (2016) Prosocial behavior mitigates the negative effects of stress in everyday life. Clin Psychol Sci 4(4):691–698. https://doi.org/10.1177/2167702615611073
Rofcanin Y, de Jong JP, Las Heras M et al. (2018) The moderating role of prosocial motivation on the association between family-supportive supervisor behaviours and employee outcomes. J Vocat Behav 107:153–167. https://doi.org/10.1016/j.jvb.2018.04.001
Stehr P (2023) The benefits of supporting others online–How online communication shapes the provision of support and its relationship with wellbeing. Comput Hum Behav 140:107568. https://doi.org/10.1016/j.chb.2022.107568
Tong Y, Tan C-H, Sia CL et al. (2022) Rural-urban healthcare access inequality challenge: Transformative roles of information technology. MIS Quart 46(4):1937–1982. https://doi.org/10.25300/MISQ/2022/14789
Wang J-J, Liu H, Cui X et al. (2023a) Impact of a physician’s prosocial behavior on the patient’s choice: an empirical investigation in online health community. Inf Technol Peopl 36(4):1703–1725. https://doi.org/10.1108/ITP-12-2020-0878
Wang Q, Qiu L, Xu W (2023b) Informal payments and doctor engagement in an online health community: An empirical investigation using generalized synthetic control. Inf Syst Res 0(0):1–21. https://doi.org/10.1287/isre.2020.475
Weinstein N, Ryan RM (2010) When helping helps: Autonomous motivation for prosocial behavior and its influence on well-being for the helper and recipient. J Pers Soc Psychol 98(2):222. https://doi.org/10.1037/a0016984
Yan Z, Kuang L, Qiu L (2022) Prosocial behaviors and economic performance: Evidence from an online mental healthcare platform. Prod Oper Manag 31(10):3859–3876. https://doi.org/10.1111/poms.13792
Yan Z, Wang T, Chen Y et al. (2016) Knowledge sharing in online health communities: A social exchange theory perspective. Inf Manag -Amster 53(5):643–653. https://doi.org/10.1016/j.im.2016.02.001
Yang H, Zhang X (2019) Investigating the effect of paid and free feedback about physicians’ telemedicine services on patients’ and physicians’ behaviors: panel data analysis. J Med Internet Res 21(3):e12156. https://doi.org/10.2196/12156
Yang Q, Huo J, Li H et al. (2023a) Can social interaction-oriented content trigger viewers’ purchasing and gift-giving behaviors? Evidence from live-streaming commerce. Internet Res 33(7):46–71. https://doi.org/10.1108/INTR-11-2021-0861
Yang Y, Zhu X, Song R et al. (2023b) Not just for the money? An examination of the motives behind physicians’ sharing of paid health information. J Inf Sci 49(1):145–163. https://doi.org/10.1177/0165551521991029
Yin Q, Guo C, He C et al. (2022) Effect of new bundled services on doctors’ online social and economic returns: Evidence from an online health community. Inf Syst Front 25:2393–2416. https://doi.org/10.1007/s10796-022-10362-6
Yoo JJ, Jhang J, Song S et al. (2023) An integrated model of prosocial crowdfunding decision: Three utility components and three informational cues. Electron Commer R A 57:101233. https://doi.org/10.1016/j.elerap.2022.101233
Zhang M, Guo X, Wu T (2019a) Impact of free contributions on private benefits in online healthcare communities. Int J Electron Comm 23(4):492–523. https://doi.org/10.1080/10864415.2019.1655208
Zhang X, Guo F, Xu T et al. (2020) What motivates physicians to share free health information on online health platforms? Inf Process Manag 57(2):102166. https://doi.org/10.1016/j.ipm.2019.102166
Zhang X, Guo X, Lai K-H et al. (2019b) How does online interactional unfairness matter for patient–doctor relationship quality in online health consultation? The contingencies of professional seniority and disease severity. Eur J Inf Syst 28(3):336–354. https://doi.org/10.1080/0960085X.2018.1547354
Zhou J, Kishore R, Amo L et al. (2022a) Description and demonstration signals as complements and substitutes in an online market for mental health care. MIS Quart 46(4):2055–2084. https://doi.org/10.25300/MISQ/2022/16122
Zhou Y, Zhu L, Wu C et al. (2022b) Do the rich grow richer? An empirical analysis of the Matthew effect in an online healthcare community. Electron Commer R A 52:101125. https://doi.org/10.1016/j.elerap.2022.101125
Zhuo X, Wang W-T (2024) Why are physicians willing to contribute knowledge? Evidence from online health communities. Comput Hum Behav 152:108095. https://doi.org/10.1016/j.chb.2023.108095
Acknowledgements
This study was supported by the National Natural Science Foundation of China (72131006).
Author information
Authors and Affiliations
Contributions
PZ: Conceptualization, Methodology, Software, Validation, Data curation, Writing—original draft, Writing—review & editing. YX: Validation, Visualization, Formal analysis, Investigation, Writing—review & editing. CL: Resources, Supervision, Project administration, Funding acquisition. JZ: Software, Investigation, Data curation. SZ: Software, Investigation, Data curation.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethical approval
This article does not contain any studies with human participants performed by any of the authors.
Informed consent
This paper does not contain any studies requiring ethical approval.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Zhou, P., Xie, Y., Liang, C. et al. The effect of prosocial behavior and its intensity on doctors’ performance in an online health community. Humanit Soc Sci Commun 11, 874 (2024). https://doi.org/10.1057/s41599-024-03420-8
Received:
Accepted:
Published:
DOI: https://doi.org/10.1057/s41599-024-03420-8
