Improving the management of hypertension and diabetes: An implementation evaluation of an electronic medical record system in Nairobi County, Kenya.
Oyugi Boniface,Makunja Sostine,Kabuti Winfred,Nyongesa Caroline,Schömburg Martin,Kibe Victor,Chege Martha,Gathu Susan,Wanyee Steven,Sahal Mohammed
International journal of medical informatics
OBJECTIVE:To evaluate the implementation of a novel electronic medical record (EMR) system for management of non-communicable diseases (NCD) (hypertension (HTN) and diabetes mellitus (DM)) in health facilities in informal settlements in Nairobi. Questions of interest were on the use of, perception of the HCWs, and scalability and sustainability of the EMR system. METHOD:The study utilised a descriptive and analytical implementation evaluation through a convergent parallel mixed-methods design in 33 health facilities in the informal settlements in Nairobi County, Kenya. We carried out semi-structured interviews with the county and sub-county health management staff (n = 9), facility in-charges (n = 8), healthcare workers (HCW) (n = 35), and project staff (n = 7). Additionally, quantitative analysis, trend analysis, critical evaluation and costing were done. Qualitative data were analysed thematically using NVIVO while quantitative data were analysed using Excel and Stata software. RESULTS:The EMR system significantly improved data capture and management of HTN and DM patients. The system helped clinicians to adhere to treatment and management guidelines and in clinical decision making. Most HCWs had a positive attitude and perceptions about the EMR system, and it was a good initiative for improving the quality and standardisation of care. The data captured made it easier to generate health facility and clinics reports which were essential for planning and decision-making processes. A critical audit of the EMR system features showed adequate general design features (data elements, structure and organisation, ease of use, accessibility, interfaces, confidentiality, access limitation, accuracy and integrity). DISCUSSION:Use of the EMR helped in improving patients care. The technology not only enhanced assurance of patients' information safety and availability but also supported in clinical decision making and standardisation of care. Successful implementation of the technology is dependent on positive perception and attitude of the HCWs. While the initial cost of setting and managing the EMR is high, future maintenance cost could be lower, making it sustainable in the long run. However, it is vital for future implementors to source for adequate funds to run it to completion if it is to achieve its objective.
An Electronic Medical Record-Based Discharge Disposition Tool Gets Bundle Busted: Decaying Relevance of Clinical Data Accuracy in Machine Learning.
Greenstein Alexander S,Teitel Jack,Mitten David J,Ricciardi Benjamin F,Myers Thomas G
Background:Determining discharge disposition after total joint arthroplasty (TJA) has been a challenge. Advances in machine learning (ML) have produced computer models that learn by example to generate predictions on future events. We hypothesized a trained ML algorithm's diagnostic accuracy will be better than that of current predictive tools to predict discharge disposition after primary TJA. Methods:This study was a retrospective cohort study from a single, tertiary referral center for primary TJA. We trained and validated an artificial neural network (ANN) based on 4368 distinct surgical encounters between 1/1/2013 and 6/28/2016. The ANN's ability to identify discharge disposition was then tested on 1452 distinct surgical encounters between 1/3/17 and 11/30/17. Results:The area under the curve and accuracy achieved during model validation were 0.973 and 91.7%, respectively, with 25% of patients being discharged to skilled nursing facilities (SNFs). Within our testing data set, 6.7% of patients went to SNFs. The performance in the testing set included an area under the curve of 0.804, accuracy of 61.3%, sensitivity of 28.9%, and specificity of 93.8%. Conclusions:This is the first prediction tool using an electronic medical record-integrated ANN to predict discharge disposition after TJA based on locally generated data. Dramatically reduced numbers of patients discharged to SNFs due to implementation of a bundled payment model lead to poor recall in the testing model. This model serves as a proof of concept for developing an ML prediction tool using a relatively small data set and subsequent integration into the electronic medical record.