Article Preview
TopIntroduction
Mental health illnesses are becoming more prevalent (Owens et al., 2019) in the United States. In 2019, NIH estimates that approximately one in five people or 51.5 million people aged 18 years and over suffered from mental and/or substance abuse disorders (MSUDs). Of these adults, nearly 45 million had a mental disorder alone, 11 million had a substance abuse disorder alone, and 8 million had both a mental disorder and a substance abuse disorder. It is further found that disorders such as depression, anxiety, and substance abuse are associated with significant distress and impairment, including complications with multiple chronic conditions, disability, inability to function in society, and substantial economic costs. Sporinova et al. (2019) cite a 3-year adjusted mean cost at $38,250 for those with a mental health disorder, and $22,280 for those without a mental health disorder. According to The American Psychological Association (Winerman, 2017), in the year 2013, $187.8 billion dollars, including out of pocket expenses, were categorized as costs related to mental disorders. Taking into account additional costs associated with loss of productivity and disability payments, the total cost of MSUDs to society is estimated to be more than twice that amount.
Hospitalization is a very important component of treatment plans for individuals with serious and persistent illness. However, hospital inpatient stay has become very expensive in countries like the USA. According to Lan Liang et. al. (2016), there were over 35 million hospital stays, equating to 104.2 stays per 100,000 population. The average cost per hospital stay was $11,700, making hospitalization one of the most expensive types of healthcare services.
According to The Piper Report (2020), hospital lengths of stay for mental health (MH) or substance abuse (SA) disorders also vary considerably, especially for mental-health related admissions. Nationwide, the MH average length of stay is 8.0 days. According to the same Report, MH and SA hospitalizations are, on average, less expensive than non-MHSA stays:
$5,700 per MH stay.
$4,600 per SA stay.
$9,300 per stay for all other conditions.
Health Catalyst, in its Newsletter issue May 25, 2017 published an article entitled “Enhancing Mental Health Care Transitions Reduces Unnecessary Costly Readmissions” and stated that “Nationally, hospitalization for persons with mental health disorders has increased faster than hospitalization for any other condition”. Also mentioned is the lack of bed space to admit the patients on a timely basis.
Therefore, it becomes necessary to assess the mental health condition of the patients. In the current study, machine learning techniques are developed, for associating the patients’ demographic, behavioral, psychological and other related data, and to evaluate the probability of hospital inpatient admission for these patients. By setting a threshold value for the probability, the medical practitioner can assess whether the patient needs inpatient admission or not. It is also interesting to assess the level of Mental Health Severity of communities, based on race, gender and patient status by using all the complex and rich data that is available. Factor analysis techniques are used here to develop a comprehensive Mental Health Severity Index (MHSI) by using the variables and to rank the communities. The rest of the article is organized in the following sections: Literature review, Materials and methods, Machine learning algorithms used, Results, and Conclusion.