The Role of Internet in Shaping the Healthcare Market Dynamics

The Internet and networking technologies have undeniably had considerable and profound influence on shaping the behavior of various entities all over the world. Furthermore, with the advent of Web 2.0, user-generated content has inundated the Internet. For example, about, 1.8 billion of photos are uploaded and shared by consumers on social media daily and 86% of businesses now use user-generated content in some way in their marketing efforts. A sizeable portion of the user-generated content relates to users' experiences with goods, services, hotels, retailers, and many other areas. As one of the major sources of information for consumers, user-generated content affects the choices made by consumers and businesses alike in a variety of ways. When such content relates to healthcare context, the consequences may be even more substantial not only from a patient-safety and patient empowerment, but also from business perspectives. For example, adverse hospital or doctor evaluations (ratings and/or reviews) might influence providers' performance outcomes as well as potential patients and even communities related to that provider. This dissertation attempts to answer the question, "What is the role of online user-generated content in shaping health care market dynamics?" Although the question can be answered in multiple ways, the focus of this dissertation is placed on the patient-provider online engagement and its benefits. Furthermore, the dissertation utilizes data pertaining to the cancer context to answer this question. The choice for using the cancer context is motivated by the fact that (a) cancer is one of the leading causes of death in the world and (b) treatment costs for cancer care very high and with the increase in use of specialty drugs the costs have and margins have increased and (c) many cases lead to loss of life. These reasons also make study more instructive from a business perspective. Three essays were developed investigating at three different levels of analysis: (1) What is the role of online user ratings in shaping the business value of oncologists? (2) How do Internet user-generated measures (UGM) of quality, awareness and favorability on social media impact the prominence of cancer-treatment hospitals? and (3) Which risk and protective indicators associated with biological, psychological, and socio-economic factors could serve as cues to detect and predict cancer incidence rates at the population (county) level? Big data analytics approach combining structured and unstructured data was employed to develop a deeper insight from the data. The first essay establishes the impact of online 5-star rating status on new patient referrals to oncologists through a prism of quadratic moderation of number of evaluations. It also provides theoretical clarifications regarding patient-provider online engagement through the physician-rating websites. The second provides conceptual clarifications regarding how each of the UGM-related factors affects prominence, while taking into account its statistical distribution in the context of hospitals. It also elucidates the implications and consequences of patient-provider engagement process in the context of online user-generated measures. Additionally, the second essay provides several managerial implications in the form of efficient, yet inexpensive solutions to increase prominence by means of Internet impression management. The third essay contributes to the medical informatics literature by (a) providing conceptual clarifications regarding the association between biopsychological factors (found in unstructured text data) and cancer incidence and (b) elucidating the role of psychological indicators (along with biological and socio-economic) in affecting county-level incidence rates while taking into account its statistical distribution. It also provides practical information that can be used by county and state officials to adjust cancer related budget allocations and overall cancer fighting strategies. It should be noted that there are significant delays before the incidence rates are collated and communicated using traditional processes. Addressing this issue, the process of surveillance using large volumes of Twitter data provides immediate visibility to the health of the community as one lens that could be used in managing community health.