Since the creation of the Division of Clinical Informatics, members of the Clinical Informatics team have published and presented innovative research. Below you can find some of their most recent work, putting our Clinical Informatics in the spotlight.
Niharika Bhardwaj, Alfred Cecchetti, Usha Murughiyan and Shirley Neitch
Caring for the growing dementia population with complex health care needs in West Virginia has been challenging due to its large, sizably rural-dwelling geriatric population and limited resource availability.
This paper aims to illustrate the application of an informatics platform to drive dementia research and quality care through a preliminary study of benzodiazepine (BZD) prescription patterns and its effects on health care use by geriatric patients.
The Maier Institute Data Mart, which contains clinical and billing data on patients aged 65 years and older (N=98,970) seen within our clinics and hospital, was created. Relevant variables were analyzed to identify BZD prescription patterns and calculate related charges and emergency department (ED) use.
Nearly one-third (4346/13,910, 31.24%) of patients with dementia received at least one BZD prescription, 20% more than those without dementia. More women than men received at least one BZD prescription. On average, patients with dementia and at least one BZD prescription sustained higher charges and visited the ED more often than those without one.
The Appalachian Informatics Platform has the potential to enhance dementia care and research through a deeper understanding of dementia, data enrichment, risk identification, and care gap analysis.
Alfred Cecchetti, Niharika Bhardwaj, Usha Murughiyan, Gouthami Kothakapu and Uma Sundaram
The Appalachian population is distinct, not just culturally and geographically but also in its health care needs, facing the most health care disparities in the United States. To meet these unique demands, Appalachian medical centers need an arsenal of analytics and data science tools with the foundation of a centralized data warehouse to transform health care data into actionable clinical interventions. However, this is an especially challenging task given the fragmented state of medical data within Appalachia and the need for integration of other types of data such as environmental, social, and economic with medical data.
This paper aims to present the structure and process of the development of an integrated platform at a midlevel Appalachian academic medical center along with its initial uses.
The Appalachian Informatics Platform was developed by the Appalachian Clinical and Translational Science Institute’s Division of Clinical Informatics and consists of 4 major components: a centralized clinical data warehouse, modeling (statistical and machine learning), visualization, and model evaluation. Data from different clinical systems, billing systems, and state- or national-level data sets were integrated into a centralized data warehouse. The platform supports research efforts by enabling curation and analysis of data using the different components, as appropriate.
The Appalachian Informatics Platform is functional and has supported several research efforts since its implementation for a variety of purposes, such as increasing knowledge of the pathophysiology of diseases, risk identification, risk prediction, and health care resource utilization research and estimation of the economic impact of diseases.
The platform provides an inexpensive yet seamless way to translate clinical and translational research ideas into clinical applications for regions similar to Appalachia that have limited resources and a largely rural population.
Niharika Bhardwaj, Shanmuga Sundaram, Larry E. Carter, Alfred A. Cecchetti and Uma Sundaram
As the burden of Metabolic syndrome (MetS) continues to rise, adults younger than 50 are developing colorectal cancer (CRC) at a steadily increasing rate. Metabolic dysfunction of MetS has been widely reported to be a significant risk factor for CRC. However, current guidelines do not consider MetS patients as high-risk merely advising the use of caution in these patients leaving them vulnerable. We explore the relationship of MetS and its individual components with CRC to determine the need for CRC screening in those with MetS and/or certain MetS components.
Materials and Methods:
We performed a retrospective analysis of de-identified patient data from Marshall University’s ACTSI Clinical Research Data Warehouse and UKHC’s Enterprise Data Warehouse between 2010 and 2018. A total of 101,757 patients aged 18 or older with all five parameters (insulin resistance, hypertension, hypertriglyceridemia, low HDL and obesity) available to classify patients as with (MetS) or without (No MetS) metabolic syndrome per the consensus definition were included in this study. Presence or Absence of CRC was determined using ICD diagnosis codes. A chi-square test of independence was performed to examine the relationship of MetS with CRC. Further, binary logistic regression was done to assess which of the MetS components or combination of MetS components was associated with a higher likelihood of CRC. Additionally, we performed a similar analysis on the population (N = 481,815) where all components weren’t measured to see whether this correlation still held true.
Our study population consisted of 99,581 females and 84,036 males with mean ages of 56.87 ± 0.055 and 57.19 ± 0.056 and mean BMI’s of 31.31 ± 0.027 and 30.18 ± 0.024, respectively. Of the 183,617 patients, 121,396 met MetS criteria while 62,221 patients did not meet MetS criteria. Within the MetS group, we found that 1608 (1.32%) had CRC while only 552 (0.89%) had CRC in the group without MetS. The relation between MetS and CRC was statistically significant (N = 183,617, p < 0.0001). Further, on regression analysis, insulin resistance was associated with the highest likelihood of CRC with patients with insulin resistance being 1.5 times more likely to have CRC than those without it (p < 0.0001). This was also seen in the population where the status of all MetS components was unknown (p < 0.0001). Other components had minimal effect, if at all.
MetS patients were more likely to have CRC than were patients without MetS. Additionally, of all the components, the presence of insulin resistance alone was associated with a higher likelihood of CRC, even in patients that did not have all the MetS parameters measured. Thus, this may indicate the need to screen patients with insulin resistance even in the absence of other MetS criteria for CRC. Further studies are warranted to understand the role of metabolic syndrome and its individual components in CRC screening.
Shanmuga Sundaram, Chris Schafer, Todd Gress and Alfred Cecchetti
Institutional data warehouse of all medical information is valuable to improve patient care, facilitate quality assurance and cost/benefit analysis. It is essential for clinical outcomes research within and between health science centers. Thus, the National Institutes of Health has made formation of an i2b2-based clinical data warehouse a key component of funding for clinical and translational science institutes in this country. Recently at Marshall a data warehouse of all clinical and billing information at Cabell Huntington Hospital and Marshall Health was created. It is currently not clear whether a warehouse may be exclusively used as source data without correlation with medical records.
Hepatitis-C patients with anemia have more GI tract pathology as etiology for the anemia.
Query from warehouse patients who had liver function tests (LFTs), hepatitis-C, anemia, and endoscopic procedures. Ascertain from medical records if the Marshall data warehouse may be used as a primary source for retrospective clinical studies. Results Between 2010 and 2015, 77,366 patients had LFTs and 37,272 were abnormal and 40,094 were normal. Hepatitis-C was more likely present in patients with abnormal LFTs (2.7 vs 0.4%). Irrespective of LFTs, Hepatitis-C patients had more anemia (75 vs 62% abnormal LFTs and 60 vs 37% normal LFTs). We will determine the presence of GI-tract pathology in each subgroup. Ultimately, we plan to determine the minimum number of randomly selected subgroup patients necessary to validate warehouse-based data with 95% confidence using the individual patient record.
The results of this study will demonstrate that warehouse may be used as primary source for clinical retrospective studies with only a small sampling validation of source data. Secondarily this study will validate the warehouse at least for this
particular patient group, which will greatly facilitate clinical research at Marshall.