City University of New York (CUNY) CUNY Academic Works All Dissertations, Theses, and Capstone Projects Dissertations, Theses, and Capstone Projects 2-2015 Risk factors and costs influencing hospitalizations due ot heat- related illnesses: patterns of hospitalization Michael T. Schmeltz Graduate Center, City University of New York How does access to this work benefit ou?y Let us know! More information about this work at: https://academicworks.cuny.edu/gc_etds/621 Discover additional works at: https://academicworks.cuny.edu This work is made publicly available by the City University of New York (CUNY). Contact: [email protected] Risk factors and costs influencing hospitalizations due to heat-related illnesses: patterns of hospitalization by Michael T. Schmeltz A dissertation submitted to the Graduate Faculty in Public Health in partial fulfillment of the requirements for the degree of Doctor of Public Health, The City University of New York 2015 © 2015 Michael T. Schmeltz All Rights Reserved ii This manuscript has been read and accepted for the Graduate Faculty in Public Health in satisfaction of the dissertation requirement for the degree of Doctor of Public Health. Jean A. Grassman ___________ ____________________________________________________ Date Chair of the Examining Committee Denis Nash ___________ ____________________________________________________ Date Executive Officer Grace Sembajwe _________________________________________________ Peter J. Marcotullio _________________________________________________ Stephanie Woolhandler _________________________________________________ David U. Himmelstein _____________________________________________________ Supervisory Committee THE CITY UNIVERSITY OF NEW YORK iii Abstract Risk factors and costs influencing hospitalizations due to heat-related illnesses: patterns of hospitalization by Michael T. Schmeltz Adviser: Professor Grace Sembajwe The objective of this dissertation was to identify individual and environmental risk factors, investigate outcomes and hospital resource use, including costs, and document the pattern of heat-related illness hospitalizations in the United States. The main data source for the study population was the 2001-2010 Nationwide Inpatient Sample (NIS). The study population for heat-related illnesses (HRIs) consists of patients in the NIS with at least one diagnosis of a heat- related illness (ICD-9 codes 992.0 – 992.9) from 2001 to 2010. Outcome analysis included a study population of patients who had primary or secondary diagnoses of diabetes, cardiovascular diseases, respiratory illnesses, nephritic illnesses and acute renal failure along with a diagnosis of a heat-related illness. Outcomes for costs were calculated and adjusted using the medical consumer price index for 2013. Data on air conditioner use and total cost of electricity use from air conditioning was derived from the Residential Energy Consumption Survey. This study identified a number of previously unknown risk factors for heat morbidity HRI, including age greater than 40, males and hospitalization in rural areas and small urban clusters. Additionally, stratified analyses of outcomes further identified specific risk factors among vulnerable populations. Elevated risk of negative health outcomes and increased hospital resource use was seen in patients diagnosed with common comorbidities, in particular those of a lower socioeconomic status, minority and most age groups with diagnoses of cardiac and respiratory iv diseases with a HRI. Analyses of costs showed substantial costs associated with hospitalizations due to heat-related illnesses with the average mean cost approximately $52.7 million while the total aggregate cost for the time period at just over half a billion dollars. Projected estimates for the average yearly cost of these hospitalizations in the future climate with estimates around half a billion US dollars by the late-21st century. In conclusion, the study revealed a number of risk factors and negative health outcomes associated with hospitalizations of heat-related illnesses. These findings provide additional scientific evidence that heat-related illnesses will continue to rise and will continue to be a public health burden as climate changes increase in frequency and intensity of extreme weather events. v Acknowledgements I want to deeply thank the members of my dissertation committee – Grace Sembajwe, Jean Grassman, Peter Marcotullio, David Himmelstein and Steffie Woolhandler – your support, guidance and expert knowledge helped me tremendously along the way and I am forever grateful. I want to especially thank Jean (chairperson) and Grace (sponsor). Jean, you were with me from day one and your unwavering support helped me navigate through this whole process successfully, thank you so much. Grace, it was perfect timing when you came to CUNY as a new faculty member and agreed to be my sponsor. Your enthusiasm, knowledge and commitment to my work buoyed me on many occasions and your excitement for my research always gave me hope that I can do this. And I did it - you are a great mentor and I cannot thank you enough. I also want to thank the professors of the Environmental and Occupational Health track, your encouragement for me, as the first EOH DPH graduate was always inspiring and supportive. I want to thank the faculty at the CUNY SPH and those I worked with specifically, Heidi Jones, Bill Gallo, Frank Mirer, Jack Caravanos, Mimi Fahs, Susan Klitzman and Trudy Lieberman. The administration of the CUNY SPH and the DPH program have always been supportive as well and helped in guiding me through the program. Thank you to the founding Dean Kenneth Olden and Dean Ayman El-Mohandes and executive officers Nick Freudenberg and Denis Nash, your personal and financial support from the school enabled me to complete the program. The administrative staff was also a big help in getting me though the program. I want to especially thank Jason Pavo and Sara Ingram. Day in and day out you answered questions, sometimes the same exact ones I asked a week prior, directed me through the bureaucracy and always listened. You gave me my (rightfully so) 15 minutes of fame in the DPH newsletter and were supportive and many ways. I will miss being in the office with both of you. It was my fellow classmates who truly understood what we were going through. We all supported and encouraged each other and I will miss that comradery and creative environment. An ‘epidemiological’ thanks to Rachael Weiss, Hannah Simons and Ellen Wiewel – this was my weakest area and you put up with my constant questions about analyses and SAS – you’re angels. My writing group – Sonia Gonzales, Amy Kwan, Liza Fuentes, Lauren Dinour and Dana Watnick – what can I say, you were the most supportive in my life during this time and we accomplished a lot together. I could not have done any of this without you. I know you will always be a part of my life and I wouldn’t want it any other way. A great thanks to my cartographer, geographer and murder mystery friend, Gretchen Culp, your guidance in spatial analysis helped me through a tough spot in my research. Finally to my close friends and family – you may not have understood why I was in school for so long, but you were always there for me no matter what. I can never thank you enough, but know that it has always meant the most to me that you continued to support me in any way you could and were always proud of me. I love you all dearly. vi Table of Contents 1. Introduction 1.1. Climate change and extreme weather events ………………………………………. 1 1.2. Heat-related illness …………………………………………………………………. 4 1.3. Vulnerable populations to extreme heat events ……………………………………. 9 1.4. Overview of the dissertation ………………….…………………………………….18 1.4.1. Overall goals …………………………………………………………….18 1.4.2. Specific aims …………………………………………………………….19 1.4.3. Organization of the dissertation ………..……………………………….20 1.4.4. Significance of the dissertation ………………………………………….21 1.5. Data sources and study population …………………………………...…………….21 2. Identifying individual risk factors and documenting the pattern of heat-related illness through hospitalization and spatial patterns of household cooling 2.1. Introduction ................................................................................................................24 2.2. Methods ......................................................................................................................27 2.2.1. Data sources .............................................................................................27 2.2.2. Subjects and definitions ............................................................................29 2.2.3. Statistical analyses ....................................................................................29 2.3. Results ........................................................................................................................30 2.3.1. Demographic and hospital characteristics ...............................................30 2.3.2. Multivariable analysis of risk factors for heat-related illness hospitalizations .........................................................................................30 2.3.3. Multilevel analysis of risk factors for heat-related illness hospitalizations .........................................................................................32 2.3.4. Spatial analysis of heat-related illness hospitalization and household cooling ..............................................................................32