Healthcare Service Supply and Demand in the Post-Pandemic World
The pandemic impacts nearly every area of life and the resulting cascading effects are likely to result in long-term consequences that require multiple resources over a long period of time. Disruption to routines, isolation from regular life experiences, limited social contact and support, coupled with uncertainty across multiple life domains contribute to decreased health and functioning. Considering most of the population’s economic lifestyle, this pandemic has adversely affected people’s financial state. The prolonged duration of this pandemic and chronic stress without a clear end period can amplify deleterious effects. The ultimate objective of this project is to provide better models for predicting healthcare needs during acute health and economic crises, particularly among economically vulnerable populations. The proposed models will help predict and assess community-based healthcare needs caused by the impact of COVID-19 on multiple domains of life, including career, family, physical health, economic security, and basic safety.
Detecting and Predicting Medical Errors in the Cancer Screening Process
Medical errors are listed among the leading causes of death in the United States and have been ranked third in 2016. These errors impose both financial and non-financial costs on patients and government. Cancer is one of the most time-consuming and costly diseases where varied preventive approaches and early detection methods have been developed by medical researchers in this area. Cancer screening is one of the most accepted means of early cancer detection prior to symptoms occurrence. Although there are controversies about the lead-time bias and survival rate in cancer patients, it’s acknowledged that screening makes the treatments easier, in some cases with even lower costs, and it can increase the chance of survival significantly. In this research we will use available data submitted in screening centers and use advanced statistical methods for analyzing operational failure in a specific screening guideline.
Analyzing Community-Based Needs for Psychiatric Disorders Using SES and Clinical Data
About 13% of emergency department visits are psychiatric patients. Patients who visited to the emergency department with mental health including substance use disorders are 2.5 times as likely to be admitted as those with purely physical problems. Overcrowding in emergency department is a major healthcare system problem due to the quality of care such as waiting times, delays to treatment, and increased costs. Mental health disparity between communities could be major issue for the quality of care such as significant demands in a specific area compared with the general population. We address the association between geo-socioeconomic status to the patterns and proximity to psychiatric inpatient healthcare services using clinical data and propose a study using a large-scale healthcare data to assess the relationship between geo-socioeconomic factors and patient outcome rate that overcomes several of the current limitations. The objective is to use analytical models to identify risk factors associated with emergency departments admission in patients with psychiatric disorders.
Crowdsourcing of Physical and Mental Health and Its Impact on Healthcare Services
Crowdsourcing refers to collection of information and opinions from a large group of people to support a specific incentive. In today’s day and age of technology and social media, access to information and reaching out to people across the world has never been easier. As well as the vast activities that can be done with the internet, many people seek advice and help through the internet. Many pieces of advice online are unprofessional but useful to connect with alike minded people across the world. This unsupervised advice and support from family, friends, and strangers has changed the dynamic of the interactions between patients and doctors. Therefore, the gathered knowledge of similar experiences of different people within the net can help ease the process of decision-making and delivering the diagnosis for the doctors. This study investigates the current relationship between doctor and patient, specifically during the diagnosis and decision-making process, to understand where, if plausible, should social media be implemented to improve the proceedings.
Data-Driven Analytics for Predicting Operational Failures in Healthcare Systems
Time-to-event datasets are the outcome of many investigations, experiments and processes in different fields such as medical science and healthcare. The increasing accessibility of information through advanced digital communications and consumers’ telecommunications devices have led to availability of massive amount of medical and healthcare data including high-dimensional covariates are increasingly accessible. Analyzing these type of data has an inevitable role in predicting the probability of many events occurrence. In such analysis, interpretability is a desirable quality. Thus, in the presence of massive and complex data, proper methods are required for efficient analysis of such data. Specifically, scoring system are models which used to assess the risk of numerous serious medical conditions since they allow users to make predictions without extensive training. This paper presents a method for failure-time data analysis in healthcare operations considering optimality, completeness, accuracy, and velocity.
* References are excluded from this narrative.