Conference PaperPDF Available

Social-Network Analysis for Pain Medications: Influential physicians may not be high-volume prescribers

Authors:

Abstract and Figures

According to the Institute of Medicine of the National Academies, more than 100 million Americans suffer from chronic pain related to diabetes, heart disease, and cancer combined. Adoption of pain medications and safe healthcare practices is a major global policy concern. This adoption process is highly influenced by the interpersonal network of physicians prescribing medications to treat pain. However, existing research into physician networks have been hospital-specific, applied to a smaller number of physicians, and dependent upon physicians' self-reports. In this paper, using big-data and data-mining, we overcome these limitations: By using a case of 30+ hospitals spanning across 2000+ physicians, we create a social network containing physicians' prescription data and adoption behavior of pain medications. The social network assumes that connected physicians work in the same hospital and belong to the same specialty or specialty group. Then, using the centrality measures, degree and eigenvector centrality, we analyze prescription volumes and proportion of adopters of pain medications. We also analyze gender effects. Results revealed that the most influential physicians were not the high-volume prescribers. Males' physicians were more influential compared to female physicians; however, females prescribed more volume compared to males. Our results help us identify critical physicians from certain core specialties and specialty groups who may be approached by patients seeking pain relief.
Content may be subject to copyright.
Social-Network Analysis for Pain Medications:
Influential physicians may not be high-volume
prescribers
Abhinav Choudhury, Shruti Kaushik, Varun Dutt
School of Computing and Electrical Engineering
Indian Institute of Technology Mandi
Himachal Pradesh, India -175005
{abhinav_choudhury, shruti_kaushik}@students.iitmandi.ac.in
varun@iitmandi.ac.in
AbstractAccording to the Institute of Medicine of the
National Academies, more than 100 million Americans suffer
from chronic pain related to diabetes, heart disease, and cancer
combined. Adoption of pain medications and safe healthcare
practices is a major global policy concern. This adoption process
is highly influenced by the interpersonal network of physicians
prescribing medications to treat pain. However, existing research
into physician networks have been hospital-specific, applied to a
smaller number of physicians, and dependent upon physicians’
self-reports. In this paper, using big-data and data-mining, we
overcome these limitations: By using a case of 30+ hospitals
spanning across 2000+ physicians, we create a social network
containing physicians prescription data and adoption behavior
of pain medications. The social network assumes that connected
physicians work in the same hospital and belong to the same
specialty or specialty group. Then, using the centrality measures,
degree and eigenvector centrality, we analyze prescription
volumes and proportion of adopters of pain medications. We also
analyze gender effects. Results revealed that the most influential
physicians were not the high-volume prescribers. Males’
physicians were more influential compared to female physicians;
however, females prescribed more volume compared to males.
Our results help us identify critical physicians from certain core
specialties and specialty groups who may be approached by
patients seeking pain relief.
KeywordsSocial network analysis; eigenvector centrality;
gender; pain medications; big-data.
I. INTRODUCTION
According to the Institute of Medicine of the National
Academies, more than 100 million Americans suffer from
chronic pain related to cases of diabetes, heart disease, and
cancer combined [1]. Pain medications are medicines
prescribed by multitude of physicians from different specialties
[2]. Adoption of pain medication is a lively process highly
dependent on the interpersonal interactions between members
of a social network.
Social Network Analysis (SNA) [3], which views the
structure of social interactions as networks composed of nodes
(physicians) interconnected by edges (social relations,
friendship, advice), is an ideal approach for describing
interaction patterns to study how social influence is transmitted
among physicians and how it affects their contingent behavior
[4]. A social network/system is the pattern of friendship,
advice, communication or support which exists among the
members of a social system [3,5-7]. Social networks can
promote innovation processes and expand opportunities for
learning [8]. Such networks can be used to find key-opinion
leaders inside a social system.
Whenever a new pain medication is released, the adoption
process in a social network is unlikely to be simultaneous, i.e.,
not all physicians adopt the medication at the same time [9].
For many new medications, an individual’s decision to adopt
the technology depends on whether his contacts have adopted it
already [10]. The adoption process may differ from one
individual to another. While some may adopt the innovation
early; others may adopt it late or may not adopt it at all [11]. In
fact, many times new medicine innovations may require long
period of time, mostly some years, from the time when they
become available in the market to the time when they maybe
widely adopted by physicians [12]. Thus, the spread or
diffusion of innovations (e.g., of pain medications) could be
thought of as asocial processthat is highly dependent on the
members of the social system. Social influence, often
crystallized as opinion exchange and behavior imitation, is
mostly conveyed through physicians interpersonal social
interactions with members of his social network [4].
Existing applications of social-network analyses for
investigation adoption of an innovation among physicians has
been hospital-specific, and were created using physicians’ self-
reports [13-14]. Thus, social-network analysis research that
considers many physicians over several hospitals in a
geographic area and that is independent of physician’s self-
reports is much needed, but lacking in literature.
In this paper, using big-data and data-mining, we create a
social network of several physicians over a number of hospitals
in the US by considering assumptions on physician’s specialty
and affiliation to the same hospital. In what follows, we first
cover background literature related to applications of social
network analyses to adoption of innovations among physicians.
Next, we build a social network and use it to draw conclusions
on critical physicians, prescription volume, and adoption
proportions.
II. RELATED WORK
The rate of adoption of a new medication (innovation)
varies from physician-to-physician [15]. Most importantly, the
medication that a physician prescribes is likely to be highly
influenced by the interpersonal communication of the
physician with members of his personal network [9]. While
some physicians may adopt a new medication early, most of
them may want to test the waters first. Thus, most physicians
may want to wait for others in his social system to have tried
the innovation first. It has been shown that physicians who are
‘socially proximate’ in a social environment often use one
another as information sources to manage the uncertainty of
adopting new antibiotic drugs [16].
Previous work has focused on questionnaire-based or
survey-based approaches to create the social network. For
example, Creswick and Westbrook [13] conducted a study in
the renal ward of an Australian metropolitan teaching hospital.
They designed a social network questionnaire containing five
social network questions about each person in the ward. The
advice-seeking interactions of doctors, nurses, allied health
professionals (including a pharmacist) and a ward clerk were
examined using the data from questionnaires administered to
staff. Keating et al. [14] have also used a survey-based
approach for creating a social network. Their study population
included an alphabetized list of 38 primary-care physicians at a
major Boston teaching hospital. Participants were asked to
“circle the number of conversations that they have had with
each of the following primary-care physicians in the past 6-
months, which have influenced their thinking on women’s
health issues” (response options were 0, 13, or 4). Physicians
were also asked to report one individual, inside or outside of
the practice, who most likely influences their practice regarding
women’s health issues.
These approaches to social network creation suffer from
sampling bias, i.e., the responses and outcomes of the survey
are highly dependent on the chosen sample. In this paper, we
create a social network by mining big-data from 30+ hospitals
and 2000+ physicians of US. Then, using the prescription data
of these physicians, we create a directed influence graph for
identifying the critical physicians using degree (in-degree and
out-degree) [3] and eigenvector centrality. Degree centrality
refers to the number of ties a physician has with other
physicians [17]. Eigenvector centrality [18] is a measure of the
influence of a physician in a network of physicians. It assigns
relative scores to nodes representing physicians in the network
based on the concept that connections to high-scoring nodes
contribute more to the score of the node in question than equal
connections to low-scoring nodes [19]. In this paper, we focus
on the diffusion of pain medications in a large network of
physicians prescribing pain medications.
III. METHOD
A. Data
We used a medical-prescriptions dataset and a physician-
affiliation dataset to create the social network of physicians
prescribing pain medications.1 These datasets are proprietary
and could be purchased from entities collecting such data
anonymously. In US, each physician is affiliated to multiple
hospitals and each hospital is affiliated to an Integrated
Delivery Network (IDN). An IDN is a formal system of
providers and hospitals that provides both healthcare services
and a health insurance plan to patients in a geographic area
[20]. We have considered an IDN located in Boston, MA,
consisting of 30+ hospitals affiliated under it and 2000+
physicians. The prescription dataset consisted of billions of
physicians prescribing certain pain medications between years
2011 and 2016.
B. Social Network Creation
The following assumption was used to create a social network:
Physicians affiliated or working in the same hospital within the
same specialty or specialty group and prescribing the same
pain medication were assumed to influence each other.
Specialty groups2 are like clusters that contain specialties from
similar fields that are related to one another. Once the social
network was created, we used the adoption pattern for a
popular pain-medication (called M) to create a directed
influence graph. The time of first prescription of the pain
medication by a physician was taken as the time of adoption of
the innovation by the physician. Physicians adopting M at an
earlier date were assumed to influence other physicians
adoption of M in his/her social network. In the directed graph,
an arrow (A, B) is directed from A to B, where ‘A’ signifies
the physician that has been influenced by physician ‘B’ for
medication’s adoption. After the creation of the social network,
we visualized the sociogram using gephi [21] and used graph
centrality measures like degree [17] and eigenvector centrality
[18] to pinpoint the critical physicians’ that have influenced the
adoption process.
IV. RESULT
In this section, we consider the results of how physicians relate
to other physicians in their network. Fig. 1 shows the directed
influence graph between various physicians in the social
network. There are 22 nodes (a node represents a physician)
and 158 edges in the social network in Figure 1. There exist
two isolated clusters in the graph belonging to different
specialty group: Primary-care physician (PCP) and NRPPHA
(Nurse-practitioner). The average path length is 1, which
indicates that, on average, there is only one tie between each
pair of nodes in the influence network. After the creation of the
social network, we applied graph centrality measures to
identify critical physicians within the network.
1 Data shown may be altered to protect privacy and uphold non-
discloser agreements.
2 These specialty groups were provided with the data.
We found that physicians from specialty internal medicine and
family medicine had the highest in-degree and eigen-vector
centralities (see Fig. 2 and 3). High in-degree centrality for a
physician signify that the physician has influenced more
number of physicians in his/her social network; while, a high
eigenvector centrality for a physician signify that the physician
is connected to highly influential nodes and is in turn also
highly influential. As such, they play a significant role in the
diffusion of the pain medication within the network.
Fig. 2. Average In-degree for different specialties.
Next, we analyzed the total number of scripts of medicine M
prescribed by each physician. We found that even though
physicians from specialties internal medicine and family
medicine were highly influential (high eigenvector centrality
and in-degree, see Fig. 2 and 3); neurology and nurse
practitioners were the ones prescribing the highest volumes of
medication M within the IDN (see Fig. 4). Thus, the most
influential physicians may not be the ones prescribing the most
medication.
Furthermore, Fig. 5 shows the proportion of physicians under
different specialty groups who had adopted the medicine M.
We observed that more number of physicians belonging to PCP
specialty were adopting the pain medication compared to
physicians from other specialty groups.
Fig. 3. Average Eigenvector centrality for different specialties.
Fig. 4. Average number of scripts prescribed for different specialty.
Fig. 5. Proportion of physicians adopting medicine M from different
specialty group.
Next, we analyzed the demographic data (gender) of the
physicians. We found that number of males (16) and female
(15) physicians prescribing the medication M were similar;
however, females (mostly nurse-practitioners) were prescribing
higher volume of scripts compared to male physicians (males:
132,004; females: 159,361). Furthermore, males had a higher
influence (high eigenvector centrality) compared to females
(males: 0.23; females: 0.05).
V. DISCUSSION AND CONCLUSION
Pain medications are the highest prescribed medication in
the US [22]; and, in this paper, using big-data and machine-
learning, we show interesting contrasting patterns between
Fig. 1. Directed Influence graph between different prescribers of
pain medication M. The numbers associated with each node are
the identification number of each physician. Two isolated clusters
of specialty group (PCP and Nurse Practitioner) are visible.
influencers, volume prescribers, and adopters of pain
medications. In this paper, we used the following assumption
to create the social network: Physicians affiliated or working in
the same hospital within the same specialty or specialty group
and prescribing the same pain medication were assumed to
influence each other. Using many physicians across several
hospitals, our results reveal that Primary-Care Physicians
(PCPs) with specialties like family medicine and internal
medicine were the most influential doctors prescribing pain
medications. However, these PCPs were not among the highest
volume prescribers of the pain medications.
Firstly, we found that physicians with specialties family
Medicine and internal Medicine were highly influential (high
eigenvector centrality) in the diffusion of pain medication
within their social network. One likely reason for this result
could be that physicians with these specialties are primary-care
providers, who are the first point-of-contact for any
undiagnosed pain issues reported by patients. These PCPs
know many other specialties as they are in the process of
regularly referring patients to these specialties.
Furthermore, the average path length was 1 in our social
network, which indicates that, on average, there is only one tie
between each pair of physician nodes in the influence network.
Thus, on average, each physician is only one step away from
each other physicians in their network and the network is very
cohesive. This unit path length is also indicative of the referral
pattern between PCPs and other medical specialties in the
network.
Second, our results showed that nurse practitioners and
neurologists were prescribing more scripts of pain medication
than other specialties. One likely reason for this observation
could be that nurse practitioners are trained to manage acute
and chronic medical conditions [23]. Thus, they come in
contact with more number of patients suffering from pain as
compared to the physicians from other specialties.
Finally, our results showed that females physicians were
prescribing more scripts of pain medication as compared to
males. Upon investigation we found that this can be attributed
to the fact that Nurse Practitioners in the IDN are mostly
females and since they are prescribing high number of scripts
of the medicine, the prescription volume is skewed towards
female physicians as compared to male physicians.
VI. LIMITATIONS
In our paper we have considered a single Integrated delivery
network (IDN) located in Massachusetts for the creating the
social network. Also we have considered only a single pain
medication for creating the directed influence graph. Even
though we started with 2000+ physicians from 30+ hospitals,
due to considering a single medication we got a smaller graph.
By using multiple pain medications we can overcome this
limitation. Also in our future work, when we will consider
multitude of IDNs’ situated all over the US and multiple pain
medications for creating the directed influence graph and also
bolster our assumptions by using data from social network
websites.
VII. FUTURE SCOPE
Social network analysis can be an efficient and powerful tool
to find key opinion leaders. In the future we will consider
referral patterns, author-co-author relationship and nomination
studies to strengthen the reliability of the connections between
the physicians. We will also take into consideration different
pain medications for creation of the directed influence
network.
ACKNOWLEDGMENT
The project was supported from grant (awards:
#IITM/CONS/PPLP/VD/03) to Varun Dutt.
REFERENCES
[1] Institute of Medicine Report from the Committee on Advancing
Pain Research, Care, and Education. 2011. Relieving Pain in
America, A Blueprint for Transforming Prevention, Care, Education
and Research. The National Academies Press
[2] Pain management" Wikipedia: The Free Encyclopedia. Wikimedia
Foundation, Inc. 21 March 2017. Web. 24 March. 2017
<https://en.wikipedia.org/wiki/Pain_management>
[3] Scott, J.: Network Analysis: A Handbook. Sage. Newbury Park, CA
(1991)
[4] Zheng, K., Padman, R., Krackhardt, D., Johnson, M. P., &
Diamond, H. S, “Social networks and physician adoption of
electronic health records: insights from an empirical study”, Journal
of the American Medical Informatics Association, 17(3), 2010: 328-
336.
[5] Knoke, D., Kuklinski, J.H., “Network analysis”. Sage. Newbury
Park, CA, 1982
[6] Burt, R.S., Minor, M.J., “Applied Network Analysis”. Sage.
Newbury Park, CA, 1983
[7] Wellman, B., “Structural analysis: From method and metaphor to
theory and substance” Contemporary Studies in Sociology. 15, 19-
61 (1997)
[8] Kolleck, Nina., "Social network analysis in innovation research:
using a mixed methods approach to analyze social innovations."
European Journal of Futures Research 1.1 2013: 1-9.6.
[9] Kaushik, Shruti, Kaushik, S., Choudhury, A., Mallik, K., Moid, A.,
& Dutt, V. "Applying Data Mining to Healthcare: A Study of Social
Network of Physicians and Patient Journeys." Machine Learning
and Data Mining in Pattern Recognition. Springer International
Publishing, 2016. 599-613.
[10] Young, H. Peyton. "The diffusion of innovations in social
networks." The economy as an evolving complex system III:
Current perspectives and future directions 267 (2006).
[11] Choudhury, A., Kaushik, S., and Dutt, V., “Data-mining
Medications for Pain: Analyzing adoption behavior of physicians
and switching behavior of patients”, Proceedings of the 2nd
International Conference on Internet of things, Data and Cloud
Computing. ACM, 2017, In press
[12] Everett Roger, “Diffusion of Innovation”, Free Press. 1983
[13] Creswick, Nerida, and Johanna I. Westbrook. "Social network
analysis of medication advice-seeking interactions among staff in an
Australian hospital." International journal of medical informatics
79.6 (2010): e116-e125.
[14] Keating, Nancy L., et al, Factors affecting influential discussions
among physicians: a social network analysis of a primary care
practice." Journal of general internal medicine 22.6 (2007): 794-
798.
[15] Valente, T.W., “Social network thresholds in the diffusion of
innovations”, Social networks. 18.1, 69-89 (1996)
[16] Coleman J, Katz E, Menzel H., “Medical innovation: a diffusion
study”, 2nd edn. New York, NY: Bobbs-Merrill, 1966.
[17] Freeman, Linton C. "Centrality in social networks conceptual
clarification." Social networks 1.3 (1978): 215-239.
[18] Ruhnau, Britta. "Eigenvector-centralitya node-centrality?" Social
networks 22.4 (2000): 357-365.
[19] Eigenvector centrality." Wikipedia: The Free Encyclopedia.
Wikimedia Foundation, Inc. 12 November 2016. Web. 24 March.
2017 <https://en.wikipedia.org/wiki/Eigenvector_centrality>
[20] Jared Landis. “Post-Acute Care Cheat Sheet: Integrated Delivery
Networks”, Advisory Board 28 April, 2014.Web. 24 March. 2017.
<https://www.advisory.com/research/post-acute-care-
collaborative/members/resources/cheat-sheets/integrated-delivery-
networks>
[21] Bastian, Mathieu, Sebastien Heymann, and Mathieu Jacomy.
"Gephi: open source software for exploring and manipulating
networks." ICWSM 8, 2009, 361-362.
[22] Leonard J. Paulozzi., Karin A. Mack., and Jason M. Hockenberry.
Vital Signs: Variation Among States in Prescribing of Opioid Pain
Relievers and Benzodiazepines United States, 2012. Morbidity
and Mortality Weekly Report, July 4, 2014, US Centers for Disease
Control, p. 564.
[23] “Nurse_practitioner." Wikipedia: The Free Encyclopedia.
Wikimedia Foundation, Inc. 18 November 2016. Web. 24 March.
2017 <https://en.wikipedia.org/wiki/Nurse_practitioner>
... With a high number of patient visits, physicians interact with more patients and, as such, can disseminate critical healthcare information swiftly and appropriately. Furthermore, as per [7], influential physicians may not be high-volume prescribers. Therefore, opinion leaders in the medical community, though maybe influential to other physicians, may not have frequent interaction with patients. ...
... Therefore, these opinion leaders did not have frequent interactions with patients. Furthermore, Choudhury, Kaushik, and Dutt [7] created a physician social network using data from 30+ hospitals in Boston containing more than 2000+ physicians and the prescription data and adoption behavior of pain medications. These authors found that influential physicians were not high-volume prescribers. ...
... Additionally, since each node has a volume ( ) attached to it, volume spread is defined as the total combined volume of the influenced nodes. Motivated from literature [7,25,26], in this paper, we have used the IC model to calculate the influence spread ( ) and volume spread (ω) of each node. ...
Chapter
Full-text available
The health crisis brought about by Covid-19 has resulted in a heightened necessity for proper and correct information dissemination to counter the prevalence of fake news and other misinformation. Doctors are the most reliable source of information regarding patients' health status, disease, treatment options, or necessary lifestyle changes. Prior research has tackled the problem of influence maximization (IM), which tries to identify the most influential physicians inside a physician's social network. However, less research has taken place on solving the problem of volume maximization (VM), which deals with finding the best set of physicians that maximize the combined volume (e.g., medicine prescribed) and influence (i.e., information disseminated). The primary objective of this work is to address the VM problem by proposing three frameworks: a reinforcement learning (RL) framework, and Instance-Based Learning (IBL), and a heuristic framework called Cost-Effective Lazy Forward (CELF)-volume algorithm, a variant of the popular CELF algorithm. We compared the proposed algorithms with a Weighted-greedy algorithm and a prefix excluding maximum influence arborescence (PMIA) IM algorithm. We used the physicianSN dataset (physician social network 181 nodes and 19,026 edges) to test different algorithms. Results revealed that the CELF-volume algorithm gave an average volume spread increment of 58% compared to the baseline algorithm but gave an average influence spread increment of only 12%. While PMIA gave the highest average influence spread increment of 46% but an average volume spread increment of only 28%. In contrast, Q-learning gave an average volume spread increment of 34% and an influence spread increment of 14%. This research highlights the utility of using reinforcement learning algorithms for finding critical physicians that can swiftly disseminate critical information to both physicians and patients.
... Thus, there may or may not exist an edge between two people; however, this edge is unlabeled, i.e., it does not have a quantifier or weight indicating the strength of the relationship between the connected people. For example, Choudhury, Kaushik, and Dutt (2017) used the binary approach for creating a social network of physicians prescribing pain medications in the US. They created a binary relationship between physicians based on the similarity of physicians' specialties and affiliations. ...
... In this section, first, we discuss the methodology for creating the social network using the binary approach. The complete description of the binary approach has been provided in a prior publication (Choudhury, Kaushik, and Dutt, 2017). Here, we discuss the main points briefly. ...
... These models considered the similarity of the physicians' attributes while creating social networks. In prior research (Choudhury, Kaushik, and Dutt, 2017), the binary approach for social network creation was proposed. In this paper, we built over prior literature to overcome the limitations of the binary approach: We proposed equal and unequal weighted approaches, which were better compared to the binary approach in accounting for the actual social ties present in the network. ...
Article
Full-text available
Understanding how relationships are structured in physician networks provides insights into how these networks influence physicians' beliefs and behaviors. This understanding would help improve strategies for disseminating medical information and guidelines. But most physician social networks mainly focus on a binary relationship where either one physician is connected or not connected to another physician without any description about the strength of the relationship. This binary relationship can lead to misinformation in the network (as acquaintances and close friends may be treated equally). In this paper, we overcome the limitation of the binary relationship by proposing a weighted influence approach among a network of physicians. A physician network is a social graph comprising of nodes (physicians) and edges between nodes (social relationships). Specifically, we attach weights to the edges to quantify the strength of the relationship between two connected physicians. In one network, we assume an un-weighted (binary) link between two connected physicians; whereas, in a second network, we assume a weighted link between two physicians. In both networks, edges are created between physicians who are affiliated with the same organization-group and affiliated or working in the same hospital within the same specialty or specialty-group. We compare both the weighted and un-weighted approaches in the network by considering the diffusion of four highly prescribed pain medications in the US. Results reveal that the weighted approach is superior compared to the un-weighted approach network in explaining the diffusion of pain medications inside the social network. Also, our results help us identify that affiliation to the same organization-group and affiliation to the same hospital are important attributes to the diffusion process. Also, weights with high values do not necessarily lead to large diffusions inside the social network. We highlight the implication of our results for the diffusion of innovations in physician networks. In: Network Modeling Analysis in Health Informatics and Bioinformatics journal.
... Understanding how the diffusion of innovation takes place inside a social network can help create strategies for better and faster dissemination of information (Choudhury, Kaushik, & Dutt, 2017). This understanding is highly relevant in the healthcare domain, where it is paramount that medical innovations and guidelines are disseminated appropriately and swiftly (Kasthuri, 2018). ...
... The number of layers was kept small so as to reduce overfitting. We also used early stopping (Chollet, 2015); i.e., training will stop when the chosen performance measure stops improving). In this paper, we have used validation loss as the performance measure for early stopping, i.e., if the validation loss stops decreasing, then the model's training is stopped. ...
Chapter
Full-text available
The diffusion of innovations (new medication or updated treatment methods) between physicians inside a social network over time (diffusion cascades) may help mitigate diseases in society. Most researchers have tried to predict the future diffusion cascade size by manually extracting graph features. However, less attention has been given to predict ing the next adopter of innovation in a diffusion cascade. In this paper, we address this literature gap by predicting the next physician in a diffusion cascade via graph embeddings. For our analyses, different graph embedding techniques and embedding dimensions were fed into a number of machine learning (ML) algorithms to predict the next physician in the cascade. DeepWalk, Node2Vec, and Structural Deep network Embedding (SDNE) algorithms were evaluated as different graph embeddings; 16, 32, and 64 were assessed as different embedding dimensions; and, multilayer perceptron (MLP) and long short-term memory (LST M) algorithms were developed for predicting the future cascades. Results indicated that SDNE with 32 dimensions gave the highest performance, followed by SDNE with 64 dimensions and SDNE with 16 dimensions. Overall, MLPs performed better compared to LST Ms. Through this research, we highlight the utility of high dimensional graph embedding algorithms and dimensions in diffusion cascade prediction in the healthcare domain. Such diffusion cascades may help rapidly spread medical innovations and guidelines among physicians.
Thesis
Networks, both in theory and practice, are increasingly central to learning, work and academic practice, and have become inseparable from most of the activities which constitute modern life. As a result, today’s Higher Education (HE) student can be viewed as having a interdependent, sociotechnical relationship with a range of people, devices, services and information resources, online and off, which they use for a range of purposes. In short, they each have their own personal learning network (PLN). Despite growing recognition of this networked paradigm, there remains a lack of understanding of what PLNs look like and how they are used, hence they remain under-represented in mainstream HE pedagogy and learning design. This could result in a disconnect between a student’s personal learning behaviours and the formal learning experience they receive from their HE institution – with potentially negative consequences for learning, engagement, student satisfaction and TEF ratings. Therefore, this research has designed, tested and deployed an innovative framework for the analysis of PLNs which captures individual PLNs, enables them to be aggregated and filtered by a range of external factors, visualised online in real-time as a series of network maps, and statistically analysed to assess the impact of a range of external shaping factors. As interdisciplinary research, concepts from Web Science, Education and the Social Sciences, as well as existing networked learning research, system modelling and social network analysis, underpin the PLN analysis framework and led to the pre-definition and generalisation of the nodes within the network. This approach meant that individual networks could be usefully aggregated, compared and analysed, thereby addressing a gap in existing networked learning research by reconciling micro (individual learning networks) and the macro (community networks) level networks for analytical purposes. The empirical data was captured from responses to an online quantitative survey which captured the nodes in the network and the type and frequency of interactions across the network. In a novel approach, this survey was embedded within a Futurelearn MOOC, extending MOOC functionality from a learning tool to a research tool. This approach to data collection provided 842 responses over eighteen months, with participants from 92 different countries, 20 different ethnicities, and ranging across the full age and demographic spectrum - making this one of the largest sample sizes for research specifically aimed at mapping learning networks to date. This research can not provide an answer to why PLNs are used in the way they are, but does provide detailed and nuanced descriptions of PLNs and, through a range of ANOVA statistical significance testing, identifies the significance of the impact of six external shaping factors on the size, usage and interaction preferences of the networks of diverse groups. The findings suggest that there is more in common between the learning networks of apparently diverse groups than the literature might suggest. Network size is significantly affected by three of the six shaping factors; network usage by four of the six (although the effect is limited as it is not universal for all PLN aspects (mode, purpose, endpoint) and subsets); however, interaction preference is much less affected and displays a considerable homogeneity across diverse sample groups and subsets. In network terms, the findings indicate potentially important changes between childhood and University at which point the network undergoes growth and development with the how, why and with whom/what learners interact all undergoing transitions. This means that HEI’s are critically positioned to guide and facilitate this network growth and development in educationally productive ways. Taken together, the findings proved informative in further developing a PLN-centred networked learning pedagogy; both confirmed and challenged aspects of the existing literature; contributed an innovative research methodology; and helped to address a gap in the networked learning field by successfully reconciling micro-level networks (egonets) and macro-level networks (wholenets) by enabling meaningful comparisons across and between diverse samples. In the future, and in the spirit of open, networked education and research, it is hoped that the framework, data collection and mapping approach can be adapted by fellow researchers for use in other contexts and domains where the focus is on understanding networks at both an individual and community level.<br/
Conference Paper
Full-text available
Providing timely healthcare is important and requires identifying critical physicians and patterns among patients' consumption of medications. However, currently, there exists little research in mining physicians' social networks and analyzing patient journeys using big-data. In this paper, using big-data and data-mining, we model the diffusion of innovation and adoption behavior of pain medications by physicians residing in the US by analyzing prescriptions. In addition, we use time-series forecasting to predict physicians' continued acceptance of pain medications. From patients' perspective, we analyze the switching and sticking behavior for two competing pain medications in the US. Our results help us identify critical physicians from certain core specialties, late adoption for certain pain medications, and success in predicting continued acceptance of pain medications. Furthermore, we find little switching and large sticking among competing medications for pain. We highlight the implications of our research for adoption behavior of physicians and patient journeys through pain.
Article
Full-text available
In 2004, the US President launched an initiative to make healthcare medical records available electronically [27]. T his initiative gives researchers an opportunity to study and mine healthcare data across hospitals, pharmacies, and physicians in order to improve the quality of care. Physicians can make better informed decisions regarding care of patients if physicians have proper understanding of patient journeys. In addition, physician healthcare decisions are influenced by their social networks. In this paper, we find patterns among patient journeys for pain medications from sickness to recovery or death. Next, we combine social network analysis and diffusion of innovation theory to analyze the diffusion patterns among physicians prescribing pain medications. Finally, we suggest an interactive visualization interface for visualizing demographic distribution of patients. T he main implication of this research is a better understanding of patient journeys via data-mining and visualizations; and, improved decision-making by physicians in treating patients.
Article
Full-text available
Background: Overprescribing of opioid pain relievers (OPR) can result in multiple adverse health outcomes, including fatal overdoses. Interstate variation in rates of prescribing OPR and other prescription drugs prone to abuse, such as benzodiazepines, might indicate areas where prescribing patterns need further evaluation. Methods: CDC analyzed a commercial database (IMS Health) to assess the potential for improved prescribing of OPR and other drugs. CDC calculated state rates and measures of variation for OPR, long-acting/extended-release (LA/ER) OPR, high-dose OPR, and benzodiazepines. Results: In 2012, prescribers wrote 82.5 OPR and 37.6 benzodiazepine prescriptions per 100 persons in the United States. State rates varied 2.7-fold for OPR and 3.7-fold for benzodiazepines. For both OPR and benzodiazepines, rates were higher in the South census region, and three Southern states were two or more standard deviations above the mean. Rates for LA/ER and high-dose OPR were highest in the Northeast. Rates varied 22-fold for one type of OPR, oxymorphone. Conclusions: Factors accounting for the regional variation are unknown. Such wide variations are unlikely to be attributable to underlying differences in the health status of the population. High rates indicate the need to identify prescribing practices that might not appropriately balance pain relief and patient safety. Implications for public health: State policy makers might reduce the harms associated with abuse of prescription drugs by implementing changes that will make the prescribing of these drugs more cautious and more consistent with clinical recommendations.
Article
Full-text available
The importance of social networks for innovation diffusion and processes of social change is widely recognized in many areas of practice and scientific disciplines. Social networks have the potential to influence learning processes, provide opportunities for problem-solving, and establish new ideas. Thus, they can foster synergy effects, bring together key resources such as know-how of participating actors, and promote innovation diffusion. There is wide agreement regarding the usefulness of empirical methods of Social Network Analysis (SNA) for innovation and futures research. Even so, studies that show the chances of implementing SNA in these fields are still missing. This contribution addresses the research gap by exploring the opportunities of a mixed methods SNA approach for innovation research. It introduces empirical results of the author’s own quantitative and qualitative investigations that concentrate on five different innovation networks in the field of Education for Sustainable Development.
Article
The intuitive background for measures of structural centrality in social networks is reviewed and existing measures are evaluated in terms of their consistency with intuitions and their interpretability.