1
|
Abstract
Purpose
For digital libraries, it is useful to understand how users search in a collection. Investigating search patterns can help them to improve the user interface, collection management and search algorithms. However, search patterns may vary widely in different parts of a collection. The purpose of this paper is to demonstrate how to identify these search patterns within a well-curated historical newspaper collection using the existing metadata.
Design/methodology/approach
The authors analyzed search logs combined with metadata records describing the content of the collection, using this metadata to create subsets in the logs corresponding to different parts of the collection.
Findings
The study shows that faceted search is more prevalent than non-faceted search in terms of number of unique queries, time spent, clicks and downloads. Distinct search patterns are observed in different parts of the collection, corresponding to historical periods, geographical regions or subject matter.
Originality/value
First, this study provides deeper insights into search behavior at a fine granularity in a historical newspaper collection, by the inclusion of the metadata in the analysis. Second, it demonstrates how to use metadata categorization as a way to analyze distinct search patterns in a collection.
Collapse
|
2
|
Islam MS, Hasan MM, Wang X, Germack HD, Noor-E-Alam M. A Systematic Review on Healthcare Analytics: Application and Theoretical Perspective of Data Mining. Healthcare (Basel) 2018; 6:E54. [PMID: 29882866 PMCID: PMC6023432 DOI: 10.3390/healthcare6020054] [Citation(s) in RCA: 66] [Impact Index Per Article: 11.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/28/2018] [Revised: 05/17/2018] [Accepted: 05/21/2018] [Indexed: 12/17/2022] Open
Abstract
The growing healthcare industry is generating a large volume of useful data on patient demographics, treatment plans, payment, and insurance coverage—attracting the attention of clinicians and scientists alike. In recent years, a number of peer-reviewed articles have addressed different dimensions of data mining application in healthcare. However, the lack of a comprehensive and systematic narrative motivated us to construct a literature review on this topic. In this paper, we present a review of the literature on healthcare analytics using data mining and big data. Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we conducted a database search between 2005 and 2016. Critical elements of the selected studies—healthcare sub-areas, data mining techniques, types of analytics, data, and data sources—were extracted to provide a systematic view of development in this field and possible future directions. We found that the existing literature mostly examines analytics in clinical and administrative decision-making. Use of human-generated data is predominant considering the wide adoption of Electronic Medical Record in clinical care. However, analytics based on website and social media data has been increasing in recent years. Lack of prescriptive analytics in practice and integration of domain expert knowledge in the decision-making process emphasizes the necessity of future research.
Collapse
Affiliation(s)
- Md Saiful Islam
- Mechanical and Industrial Engineering, Northeastern University, Boston, MA 02115, USA.
| | - Md Mahmudul Hasan
- Mechanical and Industrial Engineering, Northeastern University, Boston, MA 02115, USA.
| | - Xiaoyi Wang
- Mechanical and Industrial Engineering, Northeastern University, Boston, MA 02115, USA.
| | - Hayley D Germack
- Mechanical and Industrial Engineering, Northeastern University, Boston, MA 02115, USA.
- National Clinician Scholars Program, Yale University School of Medicine, New Haven, CT 06511, USA.
- Bouvé College of Health Sciences, Northeastern University, Boston, MA 02115, USA.
| | - Md Noor-E-Alam
- Mechanical and Industrial Engineering, Northeastern University, Boston, MA 02115, USA.
| |
Collapse
|
3
|
Cook DA, Sorensen KJ, Linderbaum JA, Pencille LJ, Rhodes DJ. Information needs of generalists and specialists using online best-practice algorithms to answer clinical questions. J Am Med Inform Assoc 2018; 24:754-761. [PMID: 28339685 DOI: 10.1093/jamia/ocx002] [Citation(s) in RCA: 12] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/06/2016] [Accepted: 12/31/2016] [Indexed: 11/13/2022] Open
Abstract
Objective To better understand clinician information needs and learning opportunities by exploring the use of best-practice algorithms across different training levels and specialties. Methods We developed interactive online algorithms (care process models [CPMs]) that integrate current guidelines, recent evidence, and local expertise to represent cross-disciplinary best practices for managing clinical problems. We reviewed CPM usage logs from January 2014 to June 2015 and compared usage across specialty and provider type. Results During the study period, 4009 clinicians (2014 physicians in practice, 1117 resident physicians, and 878 nurse practitioners/physician assistants [NP/PAs]) viewed 140 CPMs a total of 81 764 times. Usage varied from 1 to 809 views per person, and from 9 to 4615 views per CPM. Residents and NP/PAs viewed CPMs more often than practicing physicians. Among 2742 users with known specialties, generalists ( N = 1397) used CPMs more often (mean 31.8, median 7 views) than specialists ( N = 1345; mean 6.8, median 2; P < .0001). The topics used by specialists largely aligned with topics within their specialties. The top 20% of available CPMs (28/140) collectively accounted for 61% of uses. In all, 2106 clinicians (52%) returned to the same CPM more than once (average 7.8 views per topic; median 4, maximum 195). Generalists revisited topics more often than specialists (mean 8.8 vs 5.1 views per topic; P < .0001). Conclusions CPM usage varied widely across topics, specialties, and individual clinicians. Frequently viewed and recurrently viewed topics might warrant special attention. Specialists usually view topics within their specialty and may have unique information needs.
Collapse
Affiliation(s)
- David A Cook
- Knowledge Delivery Center, Mayo Clinic, Rochester, MN, USA.,Mayo Clinic Online Learning, Mayo Clinic College of Medicine, Rochester, MN, USA.,Division of General Internal Medicine, Mayo Clinic
| | | | - Jane A Linderbaum
- Knowledge Delivery Center, Mayo Clinic, Rochester, MN, USA.,Department of Cardiovascular Diseases, Mayo Clinic
| | - Laurie J Pencille
- Knowledge Delivery Center, Mayo Clinic, Rochester, MN, USA.,Center for the Science of Health Care Delivery, Mayo Clinic
| | - Deborah J Rhodes
- Knowledge Delivery Center, Mayo Clinic, Rochester, MN, USA.,Division of Preventive Medicine, Mayo Clinic
| |
Collapse
|
4
|
Daniel D, Wolbrink T, Logvinenko T, Harper M, Burns J. Association between Search Behaviors and Disease Prevalence Rates at 18 U.S. Children's Hospitals. Appl Clin Inform 2017; 8:1144-1152. [PMID: 29241251 DOI: 10.4338/aci-2017-08-ra-0137] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/23/2022] Open
Abstract
Background Usage of online resources by clinicians in training and practice can provide insight into knowledge gaps and inform development of decision support tools. Although online information seeking is often driven by encountered patient problems, the relationship between disease prevalence and search rate has not been previously characterized.
Objective This article aimed to (1) identify topics frequently searched by pediatric clinicians using UpToDate (http://www.uptodate.com) and (2) explore the association between disease prevalence rate and search rate using data from the Pediatric Health Information System.
Methods We identified the most common search queries and resources most frequently accessed on UpToDate for a cohort of 18 children's hospitals during calendar year 2012. We selected 64 of the most frequently searched diseases and matched ICD-9 data from the PHIS database during the same time period. Using linear regression, we explored the relationship between clinician query rate and disease prevalence rate.
Results The hospital cohort submitted 1,228,138 search queries across 592,454 sessions. The majority of search sessions focused on a single search topic. We identified no consistent overall association between disease prevalence and search rates. Diseases where search rate was substantially higher than prevalence rate were often infectious or immune/rheumatologic conditions, involved potentially complex diagnosis or management, and carried risk of significant morbidity or mortality. None of the examined diseases showed a decrease in search rate associated with increased disease prevalence rates.
Conclusion This is one of the first medical learning needs assessments to use large-scale, multisite data to identify topics of interest to pediatric clinicians, and to examine the relationship between disease prevalence and search rate for a set of pediatric diseases. Overall, disease search rate did not appear to be associated with hospital disease prevalence rates based on ICD-9 codes. However, some diseases were consistently searched at a higher rate than their prevalence rate; many of these diseases shared common features.
Collapse
Affiliation(s)
- Dennis Daniel
- Division of Critical Care Medicine, Department of Anesthesiology, Perioperative and Pain Medicine, Boston Children's Hospital, Boston, Massachusetts, United States.,Department of Anesthesia, Harvard Medical School, Boston, Massachusetts, United States
| | - Traci Wolbrink
- Division of Critical Care Medicine, Department of Anesthesiology, Perioperative and Pain Medicine, Boston Children's Hospital, Boston, Massachusetts, United States.,Department of Anesthesia, Harvard Medical School, Boston, Massachusetts, United States
| | - Tanya Logvinenko
- Clinical Research Center, Boston Children's Hospital, Boston, Massachusetts, United States
| | - Marvin Harper
- Department of Emergency Medicine, Boston Children's Hospital, Boston, Massachusetts, United States.,Department of Pediatrics, Harvard Medical School, Boston, Massachusetts, United States
| | - Jeffrey Burns
- Division of Critical Care Medicine, Department of Anesthesiology, Perioperative and Pain Medicine, Boston Children's Hospital, Boston, Massachusetts, United States.,Department of Anesthesia, Harvard Medical School, Boston, Massachusetts, United States
| |
Collapse
|
5
|
Valtis YK, Rosenberg J, Bhandari S, Wachter K, Teichman M, Beauvais S, Weintraub R. Evidence-based medicine for all: what we can learn from a programme providing free access to an online clinical resource to health workers in resource-limited settings. BMJ Glob Health 2017; 1:e000041. [PMID: 28588926 PMCID: PMC5321332 DOI: 10.1136/bmjgh-2016-000041] [Citation(s) in RCA: 8] [Impact Index Per Article: 1.1] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/14/2016] [Revised: 04/03/2016] [Accepted: 04/21/2016] [Indexed: 11/04/2022] Open
Abstract
The rapidly changing landscape of medical knowledge and guidelines requires health professionals to have immediate access to current, reliable clinical resources. Access to evidence is instrumental in reducing diagnostic errors and generating better health outcomes. UpToDate, a leading evidence-based clinical resource is used extensively in the USA and other regions of the world and has been linked to lower mortality and length of stay in US hospitals. In 2009, the Global Health Delivery Project collaborated with UpToDate to provide free subscriptions to qualifying health workers in resource-limited settings. We evaluated the provision of UpToDate access to health workers by analysing their usage patterns. Since 2009, ∼2000 individual physicians and healthcare institutions from 116 countries have received free access to UpToDate through our programme. During 2013-2014, users logged into UpToDate ∼150 000 times; 61% of users logged in at least weekly; users in Africa were responsible for 54% of the total usage. Search patterns reflected local epidemiology with 'clinical manifestations of malaria' as the top search in Africa, and 'management of hepatitis B' as the top search in Asia. Our programme demonstrates that there are barriers to evidence-based clinical knowledge in resource-limited settings we can help remove. Some assumed barriers to its expansion (poor internet connectivity, lack of training and infrastructure) might pose less of a burden than subscription fees.
Collapse
Affiliation(s)
- Yannis K Valtis
- Harvard Medical School, Boston, Massachusetts, USA.,Global Health Delivery Project at Harvard University, Cambridge, Massachusetts, USA
| | - Julie Rosenberg
- Global Health Delivery Project at Harvard University, Cambridge, Massachusetts, USA
| | - Sudip Bhandari
- Global Health Delivery Project at Harvard University, Cambridge, Massachusetts, USA
| | - Keri Wachter
- Harvard Medical School, Boston, Massachusetts, USA.,Global Health Delivery Project at Harvard University, Cambridge, Massachusetts, USA
| | - Marie Teichman
- Global Health Delivery Project at Harvard University, Cambridge, Massachusetts, USA
| | - Sophie Beauvais
- Global Health Delivery Project at Harvard University, Cambridge, Massachusetts, USA
| | - Rebecca Weintraub
- Harvard Medical School, Boston, Massachusetts, USA.,Global Health Delivery Project at Harvard University, Cambridge, Massachusetts, USA.,Brigham and Women's Hospital, Boston, Massachusetts, USA
| |
Collapse
|
6
|
Hsieh RW, Chen L, Chen TF, Liang JC, Lin TB, Chen YY, Tsai CC. The Association Between Internet Use and Ambulatory Care-Seeking Behaviors in Taiwan: A Cross-Sectional Study. J Med Internet Res 2016; 18:e319. [PMID: 27927606 PMCID: PMC5177739 DOI: 10.2196/jmir.5498] [Citation(s) in RCA: 21] [Impact Index Per Article: 2.6] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/02/2016] [Revised: 06/29/2016] [Accepted: 11/23/2016] [Indexed: 11/21/2022] Open
Abstract
Background Compared with the traditional ways of gaining health-related information from newspapers, magazines, radio, and television, the Internet is inexpensive, accessible, and conveys diverse opinions. Several studies on how increasing Internet use affected outpatient clinic visits were inconclusive. Objective The objective of this study was to examine the role of Internet use on ambulatory care-seeking behaviors as indicated by the number of outpatient clinic visits after adjusting for confounding variables. Methods We conducted this study using a sample randomly selected from the general population in Taiwan. To handle the missing data, we built a multivariate logistic regression model for propensity score matching using age and sex as the independent variables. The questionnaires with no missing data were then included in a multivariate linear regression model for examining the association between Internet use and outpatient clinic visits. Results We included a sample of 293 participants who answered the questionnaire with no missing data in the multivariate linear regression model. We found that Internet use was significantly associated with more outpatient clinic visits (P=.04). The participants with chronic diseases tended to make more outpatient clinic visits (P<.01). Conclusions The inconsistent quality of health-related information obtained from the Internet may be associated with patients’ increasing need for interpreting and discussing the information with health care professionals, thus resulting in an increasing number of outpatient clinic visits. In addition, the media literacy of Web-based health-related information seekers may also affect their ambulatory care-seeking behaviors, such as outpatient clinic visits.
Collapse
Affiliation(s)
- Ronan Wenhan Hsieh
- Department of Medicine, National Taiwan University College of Medicine, Taipei, Taiwan
| | - Likwang Chen
- Institute of Population Health Sciences, National Health Research Institutes, Miao-Li, Taiwan
| | - Tsung-Fu Chen
- National Taiwan University College of Law, Taipei, Taiwan
| | - Jyh-Chong Liang
- Graduate Institute of Digital Learning and Education, National Taiwan University of Science and Technology, Taipei, Taiwan
| | - Tzu-Bin Lin
- Department of Education, National Taiwan Normal University, Taipei, Taiwan
| | - Yen-Yuan Chen
- Graduate Institute of Medical Education & Bioethics, National Taiwan University College of Medicine, Taipei, Taiwan
| | - Chin-Chung Tsai
- Graduate Institute of Digital Learning and Education, National Taiwan University of Science and Technology, Taipei, Taiwan
| |
Collapse
|
7
|
Agarwal V, Zhang L, Zhu J, Fang S, Cheng T, Hong C, Shah NH. Impact of Predicting Health Care Utilization Via Web Search Behavior: A Data-Driven Analysis. J Med Internet Res 2016; 18:e251. [PMID: 27655225 PMCID: PMC5052461 DOI: 10.2196/jmir.6240] [Citation(s) in RCA: 28] [Impact Index Per Article: 3.5] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/20/2016] [Revised: 07/26/2016] [Accepted: 07/27/2016] [Indexed: 01/12/2023] Open
Abstract
BACKGROUND By recent estimates, the steady rise in health care costs has deprived more than 45 million Americans of health care services and has encouraged health care providers to better understand the key drivers of health care utilization from a population health management perspective. Prior studies suggest the feasibility of mining population-level patterns of health care resource utilization from observational analysis of Internet search logs; however, the utility of the endeavor to the various stakeholders in a health ecosystem remains unclear. OBJECTIVE The aim was to carry out a closed-loop evaluation of the utility of health care use predictions using the conversion rates of advertisements that were displayed to the predicted future utilizers as a surrogate. The statistical models to predict the probability of user's future visit to a medical facility were built using effective predictors of health care resource utilization, extracted from a deidentified dataset of geotagged mobile Internet search logs representing searches made by users of the Baidu search engine between March 2015 and May 2015. METHODS We inferred presence within the geofence of a medical facility from location and duration information from users' search logs and putatively assigned medical facility visit labels to qualifying search logs. We constructed a matrix of general, semantic, and location-based features from search logs of users that had 42 or more search days preceding a medical facility visit as well as from search logs of users that had no medical visits and trained statistical learners for predicting future medical visits. We then carried out a closed-loop evaluation of the utility of health care use predictions using the show conversion rates of advertisements displayed to the predicted future utilizers. In the context of behaviorally targeted advertising, wherein health care providers are interested in minimizing their cost per conversion, the association between show conversion rate and predicted utilization score, served as a surrogate measure of the model's utility. RESULTS We obtained the highest area under the curve (0.796) in medical visit prediction with our random forests model and daywise features. Ablating feature categories one at a time showed that the model performance worsened the most when location features were dropped. An online evaluation in which advertisements were served to users who had a high predicted probability of a future medical visit showed a 3.96% increase in the show conversion rate. CONCLUSIONS Results from our experiments done in a research setting suggest that it is possible to accurately predict future patient visits from geotagged mobile search logs. Results from the offline and online experiments on the utility of health utilization predictions suggest that such prediction can have utility for health care providers.
Collapse
Affiliation(s)
- Vibhu Agarwal
- Biomedical Informatics Training Program, Stanford University, Stanford, CA, United States.
| | | | | | | | | | | | | |
Collapse
|
8
|
Thorner AR, Cao B, Jiang T, Warner AJ, Bonis PA. Correlation Between UpToDate Searches and Reported Cases of Middle East Respiratory Syndrome During Outbreaks in Saudi Arabia. Open Forum Infect Dis 2016; 3:ofw043. [PMID: 27011953 PMCID: PMC4803184 DOI: 10.1093/ofid/ofw043] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.4] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/04/2016] [Accepted: 02/15/2016] [Indexed: 11/16/2022] Open
Abstract
UpToDate search activity is useful for detecting and monitoring outbreaks of Middle East respiratory syndrome in Saudi Arabia. Background. UpToDate is an online clinical decision support resource that is used extensively by clinicians around the world. Digital surveillance techniques have shown promise to aid with the detection and monitoring of infectious disease outbreaks. We sought to determine whether UpToDate searches for Middle East respiratory syndrome (MERS) could be used to detect and monitor MERS outbreaks in Saudi Arabia. Methods. We analyzed daily searches related to MERS in Jeddah and Riyadh, Saudi Arabia during 3 outbreaks in these cities in 2014 and 2015 and compared them with reported cases during the same periods. We also compared UpToDate MERS searches in the affected cities to those in a composite of 4 negative control cities for the 2 outbreaks in 2014. Results. UpToDate MERS searches during all 3 MERS outbreaks in Saudi Arabia showed a correlation to reported cases. In addition, UpToDate MERS search volume in Jeddah and Riyadh during the outbreak periods in 2014 was significantly higher than the concurrent search volume in the 4 negative control cities. In contrast, during the baseline periods, there was no difference between UpToDate searches for MERS in the affected cities compared with the negative control cities. Conclusions. UpToDate search activity seems to be useful for detecting and monitoring outbreaks of MERS in Saudi Arabia.
Collapse
Affiliation(s)
- Anna R Thorner
- UpToDate, Wolters Kluwer Health, Waltham; Brigham and Women's Hospital, Boston; Dana-Farber Cancer Institute, Boston; Harvard Medical School, Boston
| | - Bin Cao
- UpToDate, Wolters Kluwer Health , Waltham
| | | | | | - Peter A Bonis
- UpToDate, Wolters Kluwer Health, Waltham; Tufts University School of Medicine, Boston, Massachusetts
| |
Collapse
|