CC BY
7
Section 2. Mathematical and instrumental methods of economics
https://doi.org/10.29013/EJEMS-20-2-7-13
Wang Yuewei, Shanghai Qibao Dwight High School E-mail: wangyuwei_sara@163.com
FINANCIAL WORRIES OVER MEDICAL COST AMONG ADULTS IN2017
Abstract:
Aim: This study aims to 1) examine the predictors of adults' financial worries over medical cost in 2017 2) build a predictive model for adults' financial worries over medical costusing artificial neural network and compare its performance to logistic regression model.
Method: The National Health Interview Survey (NHIS) in 2017 was used. All the eligible participants were randomly assigned into 2 groups: training sample and testing sample. Two models were built using training sample: artificial neural network and logistic regression. Receiver operating characteristic (ROC) were calculated and compared for these two models for their discrimination capability.
Results: About 26.4% of 26031 Adults had Financial worries over medical cost, about 28.2% among the female and 24.3% among the male.
According to the logistic regression, the male was 19.6% less likely than the female to have financial worries over medical cost. The non-Hispanic adults were 55.2% less likely to have financial worries over medical cost than Hispanic adults. Married people were 12.3% less worried. White were 10.7% less worried and Black population were 31.8% more worried. Compared to residents in Northeast, people in Midwest (12.0%), South (28.0%) and West (11.0%) were more worried about the medical cost. Compared to people who were not employed but looking, people who were employed (29.4%) or not employed and not looking (46.5%) were less worried. According to this neural network, the most important predictors was age, sex, working status and race.
For training sample, the ROC was 0.61 for the Logistic regression and 0.67 for the artificial neural network. In testing sample, the ROC was 0.60 for the Logistic regression and 0.65 for the artificial neural network.
Conclusions: In this study, we identified sseveral important predictors for parents' financial worries over medical cost in 2017 e.g., age, gender, race and working status. The findings can help identify people at higher risk of having the financial worries over medical cost.
Keywords: finance, medical cost, America, Logistic Regression.
1. Introduction: Administrative costs, drug costs, defensive medicine, expensive mix of treatment and other components which makes medical costs a heavy burden over people's shoulders.
According to the most recent data available from the Centers for Medicare and Medicaid Services (CMS), "the average American spent $9,596 on healthcare" in 2012, which was "up significantly from $7,700 in 2007."It was also more than twice the per capita average of other developed nations, but still, in 2015, experts predicted continued sharp increases: "Health care spending per person is expected to surpass $10,000 in 2016 and then march steadily higher to $14,944 in 2023" (electronic source - Here Are Americans' Top Financial Concerns) [1-4].
American adults' biggest financial worry is the inability to pay the medical costs in the event of a serious illness or accident, reports Gallup in new survey data. A majority (54%) of the more than 1,000 US adults surveyed said they're either very (30%) or moderately (24%) worried about this [5-8].
This study aims to 1) examine the predictors of adults' financial worries over medical cost in 2017; 2) build a predictive model for adults' financial worries over medical cost using artificial neural network and compare its performance to logistic regression model.
2. Data and Methods:
Data:
The National Health Interview Survey (NHIS) is the principal source of information on the health of the civilian noninstitutionalized population of the United States and is one of the major data collection programs of the National Center for Health Statistics (NCHS) which is part of the Centers for Disease Control and Prevention (CDC).
The National Health Interview Survey (NHIS) Data 2017 was used in this study.
Models:
We used logistic regression models to calculate the predicted risk. Logistic regression is a part of a category of statistical models called generalized linear models, and it allows one to predict a discrete
outcome from a set of variables that may be continuous, discrete, dichotomous, or a combination of these. Typically, the dependent variable is dichoto-mous and the independent variables are either categorical or continuous.
The logistic regression model can be expressed with the formula:
In (P-i ) = A0 + A • X1 + p2 • X2 +... + pn• x„
A package called "neuralnet" in R was used to conduct neural network analysis. The package neuralnet focuses on multi-layer perceptrons (MLP, Bishop, 1995), which are well applicable when modeling functional relationships.
Variables:
The outcome variable is percentage of how worried are you right now about not having enough money for Medical Cost? (ASIRETR)
Table 1.- Variables used in this study
SEX 1: male
2: female
ORIGIN_I Hispanic Ethnicity:
1: yes; 2: no
RACRECI3 1: White
2: Black
3: Asian
4: All other race groups*
AGE_P Age <18 years old
0-17
Region 1 Northeast
2 Midwest
3 South
4 West
3. Results
About 26.4% of26031 Adults had financial worries over Medical Cost, about 28.2% among the female and 24.3% among the male.
Basically, a corrgram is a graphical representation of the cells of a matrix of correlations. The idea is to display the pattern ofcorrelationsin terms oftheir signs
and magnitudes using visual thinning and correlation- value. The positive correlations are shown in blue, while based variable ordering. Moreover, the cells of the ma- the negative correlationsare shown in red; the darker trix can be shaded or colored to show the correlation the hue, the greater the magnitude of the correlation.
Financial Worries Over Medical Cost
WodcdCan_*orr)'
Figure 1. Matrix of correlations between variables Table 2.- Logistic Regression for Having Financial worries over Medical Cost
Estimate Std. Error z value Pr(> z )
(Intercept) 0.156 0.097 1.600 0.110
AGE P 0.000 0.001 0.107 0.915
Male -0.219 0.029 -7.519 0.000 ***
HISPAN NO -0.802 0.041 -19.613 0.000 ***
MARRIED -0.131 0.029 -4.456 0.000 ***
White -0.113 0.056 -2.017 0.044 *
Black 0.276 0.068 4.075 0.000 ***
Midwest 0.113 0.048 2.377 0.017 *
South 0.247 0.044 5.635 0.000 ***
West 0.104 0.048 2.167 0.030 *
Working -0.348 0.058 -6.025 0.000 ***
NWorNLor -0.625 0.062 -10.128 0.000 ***
According to the logistic regression, the male was 19.6% less likely than the female to have financial worries over medical cost. The non-Hispanic adults were 55.2% less likely to have financial worries over medical cost than Hispanic adults. Married people were 12.3% less worried. White were 10.7% less worried and Black population were 31.8% more
worried. Compared to residents in Northeast, people in Midwest (12.0%), South (28.0%) and West (11.0%) were more worried about the medical cost. Compared to people who were not employed but looking, people who were employed (29.4%) or not employed and not looking (46.5%) were less worried.
Table 2 a.- Odds Ratio and Risk Increase Based on the Logistic Regression
Estimate Odds Ratio Risk Increase
(Intercept) 0.156 116.9% 16.9%
AGE P 0.000 100.0% 0.0%
Male -0.219 80.4% -19.6%
HISPAN NO -0.802 44.8% -55.2%
MARRIED -0.131 87.7% -12.3%
White -0.113 89.3% -10.7%
Black 0.276 131.8% 31.8%
Midwest 0.113 112.0% 12.0%
South 0.247 128.0% 28.0%
West 0.104 111.0% 11.0%
Working -0.348 70.6% -29.4%
NWorNLor -0.625 53.5% -46.5%
0,0% 50,0%
I Risk Increase ■ Odds Ratio
Figure 2. Artificial Neural Network in training sample
Figure 3. Variable Importance in Artificial Neural Network
In above plot, line thickness represents weight training algorithm has converged and therefore the magnitude and line color weight sign (black = posi- model is ready to be used.
tive, grey = negative). The net is essentially a black box so we cannot say that much about the fitting, the weights and the model. Suffice to say that the
According to this neural network, the most important predictors were age, sex, working status and race.
Fugire 4.
For training sample, the ROC was 0.61 for the the Logistic regression and 0.65 for the artificial Logistic regression and 0.67 for the artificial neural neural network. network. In testing sample, the ROC was 0.60 for
Figure 4. ROC in training sample for Logistic Regression (Red) vs Neural Network (Blue)
Figure 5. ROC in testing sample for Logistic Regression (Red) vs Neural Network (Blue)
4. Discussion
This study aimed to 1) examine the predictors of adults' financial worries over Medical Cost in 2017; 2) build a predictive model for adults' financial worries over Medical Cost using artificial neural network and compare its performance to logistic regression model.
About 26.4% of26031 Adults had Financial worries over Medical Cost, about 28.2% among the female and 24.3% among the male.
According to the logistic regression, the male was 19.6% less likely than the female to have financial worries over medical cost. The non-Hispanic adults were 55.2% less likely to have financial worries over medical cost than Hispanic adults. Married people were 12.3% less worried. White were 10.7% less worried and Black population were 31.8% more worried. Compared to residents in Northeast, people in Midwest (12.0%), South (28.0%) and West (11.0%) were more worried about the medical cost. Compared to people who were not employed but looking, people who were employed (29.4%) or not employed and not looking (46.5%) were less
worried. According to this neural network, the most important predictors was age, sex, working status and race.
For training sample, the ROC was 0.61 for the Logistic regression and 0.67 for the artificial neural network. In testing sample, the ROC was 0.60 for the Logistic regression and 0.65 for the artificial neural network.
There are limitations in this study. For example, we did not include the health conditions in this study when examing the factors of the financial worries of the medical cost. Health status could be a important factor for this topic and should be included in the future analysis when feasible.
In this study, we identified several important predictors for parents' financial worries over medical Cost in 2017 e.g., age, gender, race and working status. We built a predictive model using artificial neural network as well as logistic regression to provide a tool for early detection.
This research can provide the base for further research as how medical cost affects people and how to deal with high medical cost.
References:
1. Here Are Americans' Top Financial Concerns // URL: https://www.marketingcharts.com/industries/ education-77468 (Electronic source. Accessed date: 07.23.019).
2. Ester Bloom. Here's how much the average American spends on health care // URL: https://www. cnbc.com/2017/06/23/heres-how-much-the-average-american-spends-on-health-care.html (Electronic source. Accessed date: 07.23.019).
3. About the National Health Interview Survey // URL: https://www.cdc.gov/nchs/nhis/about_nhis. htm (Electronic resources. Accessed date: 07.23.019).
4. Parker Kim, Horowitz, Juliana Menasce. Parenting in America: Outlook, Worries, Aspirations are Strongly Linked to Financial Situation // Center for victim research repository. 2015.- Dec. 17-202. 419. 4372.
5. Andrew Wroe. Economic Insecurity and Political Trust in the United States // American Politics Research. 2015.- Aug. 7.
6. Xiaochen Ding. Research on the development of American health industry // Jilin University -2018-F719.
7. Rui Jiao. The establishment and reform of medical care in the United States // Suzhou University-2016-F847.12.
8. Lina Ren. A new political economy study on the evolution of American medical insurance system // Liaoning University-2019-R197.1; F842. 684.
