Does Covid 19 Pandemic Have a Direct Impact on the Level of Trust and Commitment toward Hospitals as Health Service Providers in Yogyakarta ?

The purpose of this study was to examine the direct and indirect effects of the pandemic on consumer trust and consumer commitment towards public health care in Yogyakarta during the pandemic. The research method in this study was the survey method with path analysis tools. Based on the research result it can be concluded that the direct impact of the Covid pandemic (X) on trust (Y1) is 0.205 (unstandardized) and 0.187 (standardized), the direct impact of trust (Y1) on the Y2 commitment is 0.602 (unstandardized) and 0.63 (standardized). Based on the ttest on the Sobel formula, the t value (3.251) > t table (1.663), means that there is an indirect impact between variable X (pandemic) on Y2 (commitment) through Y1 (trust). The impact of the Covid pandemic indirectly affects consumer commitment through the variable of trust with a combined direct and indirect impact of 1,099.


INTRODUCTION
Indonesia has been hit by a catastrophic Covid 19 outbreak since March 2020 The last phase is a phase where the hospital financial crisis may occur because a lot of operational personnel were drained in the previous phase. This happened in the East Java Government, which ran out of budget to finance the operations of a special hospital for Covid-19 patients due to the large expenditure for mitigating the impact of pandemic. The average cost of care per Covid-19 patient is IDR 50 million. Some hospitals take a policy by laying off their employees. Faisal Islamic Hospital in Makassar City has temporarily suspended their 157 employees because the number of regular patients has decreased significantly by around 80-90%. Furthermore, 80 volunteers at RSUD Lasinrang Pinrang also had to be laid off because the number of patients was reduced by 70% compared to the normal situations. Most of the biggest national hospital companies also felt the impact of this pandemic. Of the six hospital group listed on the Indonesia Stock Exchange during the period January 2-May 15 2020, all of their share prices plummeted. Omni Hospital's shares fell 51.26%, Mayapada Hospital shrank 45.37%, Royal Prima Hospital fell 35.26%, Hermina Hospital fell 20.98%, Siloam Hospital fell 20.28%, and Mitra Keluarga Hospital also fell 12, 77%. ( sumber: https://nationalgeographic.grid.id/, 2020).
Consumer trust in the local health system and hospitals and their technological capabilities is urgently needed by hospitals in times of a pandemic, as many consumers tend to avoid hospitals temporarily. Health systems and hospitals have several opportunities to cultivate this trust and improve their relationships with consumers in ways that will help them cope with the current crisis and its long-term impact.
In today's digitalization era, health care providers can foster consumer trust by communicating, providing information and proactive guidance to communicate with patients.
Hospitals should provide alternative solutions such as telephone consultations and virtual interactions such as webinars, virtual consultation or free online health education as a new way of providing health services. The hospitals also need to make it comfortable when patients arrive in person, establish the safety, hygiene and social distancing measures implemented by health institutions. The rapid change in consumer behavior pushes the industry to quickly respond to consumers with telehealth and telemedicine.
Changes in consumer behavior from before the pandemic and during the pandemic toward the health care providers such as hospitals prompted the authors to analyze the level of consumer trust and commitment to hospitals today. People tend to postpone their routine health checks schedule compared to pre-pandemic. This trend needs to be analyzed as a consideration for hospitals to improve service systems that prioritize empathy to increase trust during a pandemic. Trust and commitment are factors that influence the consumer views and behavior in using services. Research questions: (1) does the COVID-19 pandemic have a direct impact on the level of consumer trust and commitment to hospitals as health service providers?; (2) Does the COVID-19 pandemic have an indirect impact on the level of consumer trust and commitment to hospitals as health service providers?
The purpose of this study was to identify whether the Covid 19 pandemic had a direct and indirect impact on the level of consumer trust and commitment to hospitals as health care providers.

METHODS
Research design is an investigation plan and structure designed to obtain answers to the research questions. The research design in this study used a descriptive method. The purpose of descriptive analysis is to describe the characteristics of several variables in a particular phenomenon (Sekaran and Bougie, 2013: 97).
The population is the whole set of elements that are the object of research that the researcher uses to draw conclusions (Cooper and Schindler, 2011: 364). The populations in this study were hospital service users in two district areas, Ngemplak, Sleman during the period October-December 2020. The sampling technique is to determine how to take a sample from the population that will be used in the study. This study uses probability sampling techniques with simple random sampling to provide equal opportunities for each population to be randomly selected regardless of the existing strata in the population. (Sugiyono, 2018). The sampling technique in this study was carried out by purposive sampling, namely consumers who use whatsapp, whatsapp group or email to make it easier to send questionnaires while a small number use manual or offline methods.
From the links that have been distributed in the two hamlets, there was 90 samples collected and the number of appropriate samples in this study was 86 respondents. Determination of sample size is based on calculations and reference tables developed by experts. Roscoe (1975) quoted by Uma Sekaran (Sekaran and Bougie, 2013: 269) provides a general reference for determining an appropriate sample size which is more than 30 and less than 500 for most studies.
Taking the number of samples can be determined if the population is known. The population of the two sampling areas in two district areas was 600 people. Then the sample calculation can use the Yamane and Isaac and Michael formula as follows: n= N N.d²+1 Dimana: n = Number of sample N = Number of population d²= Designated precision n= 600 600.(0.1)²+1 = 85.71429 Therefore the number of samples is determined to be at least 85 people.
The variables used in this study are: 1. Dependent Variable. The dependent variable is a variable that is influenced or a result of the independent variable (Sugiyono, 2009: 59). Independent variables are variables that affect or cause changes or the emergence of the dependent variable (Sugiyono, 2009: 59).
1. Commitment (Y1). Customer commitment is defined as an attitude or desire to use a particular brand or company (Moorman et al, 1992). Based on the research of Fullerton (2003) and Hansen et al (2003) in (Gustafsson et al., 2005), the definitions of commitment are divided into two dimensions, namely affective commitment and calculative commitment. The calculative commitment is more rational, taking into account the economic benefits and costs incurred for obtaining the goods. Meanwhile, affective commitment is more emotional based on personal feeling involvement. Effective commitment is formed when the customer feels comfortable. 2. Consumer Trust (Y2). According to Felix Gille, Sarah Smith and Nicholas Mays (2015), trust is very important in health care in health services. A clear example of the value of trust lies in its effect on the level of adherence to health procedures that are being carried out. Moreover, it recognizes that enhancing and justifying trust in health care policies is embedded as a fundamental ethical goal. Lack of trust may have harmful effects on the patient's health by delaying the patient's consultation with the doctor or the absence of necessary patient information. Customers can get positive and negative information from various sources and the internet can help to disseminate information about trust and satisfaction, including users of certain hospital services. 3. The Covid Pandemic (X). The Covid 19 outbreak is a problem that affects the world today. The epidemic has an impact in various fields ranging from economic, business, social. Changes in consumer behavior occur in various industrial sectors including tourism, retail, education and health services (Naveen, D;Gustafsson, 2020).The use of the internet and social media is the main choice in social interaction today. In many cases nowadays the internet is the main way to get essential services including appointments with a doctor. The Covid outbreak is likely to lead to bankruptcy for many well-known brands in many industries due to changes in consumer behavior to stay at home due to regulatory requirements (Tucker in Naveen, D; Gustafsson, 2020) and also from the level of fear of the virus itself.

RESULTS AND DISCUSSION
The purpose of Ghozali's (2016) multicollinearity test is to identify whether the questionnaire items are related to one another. One method used to detect the presence or absence of multicollinearity is to test the tolerance and Variance Inflation Factor (VIF).By measuring the tolerance value of variability, a low tolerance value means a high VIF value, with a tolerance value of 0.10 or a VIF value above the number 10. From table 1.6, it can be seen that the VIF value is <10. The table shows that there is no VIF (variance inflation factor) value above 10, it can be concluded that there are no multicollinearity symptoms in this research equation. There is no high correlation between the two independent variables in this study so that the relationship between the independent variables and the dependent variable is not biased. So it can be concluded that there is no correlation between the independent variables and the regression model fulfills the multicollinearity assumption. The multicollinearity test aims to determine whether there is a high relationship between the independent variables. There is no multicollinearity problem if the tolerance value is above 0.1 and VIF is below 10. The results above show that the Tolerance value of each independent variable is above 0.1 and VIF is below 10 so it can be said that there is no multicollinearity. This shows that there is no high relationship between the independent variables.

Variable relationship model
In this study, X is the Pandemic as an independent variable, while Y1 is Trust and Y2 is Commitment as the dependent variable. The hypotheses in this study are: Ho: The impact of the Covid pandemic directly on the level of consumer trust and commitment Ha: The impact of the Covid pandemic indirectly on consumer commitment through trust variables Equations: Y1= a+bX Y2= a+bX+bY1 The equations for calculating the indirect impact through the trust variable: Direct impact of X (Impact of the Covid pandemic) on Y2 (commitment) =P1 indirect impact of X (impact of the covid pandemic) on Y2 (commitment) through Y1 (trust) The sum of direct and indirect impact = P1+ (P2xP3) Correlation coefficient is a measure of the value between two variables. The magnitude of the correlation coefficient ranges from -1 to +1. Correlation coefficient shows the strength of the linear relationship and the direction of the relationship between two random variables. If the correlation coefficient is positive, then the two variables have a unidirectional relationship. To make it easier to interpret the strength of the relationship between the two variables, the author provides the following description (Sarwono: 2006): • 0 : There is no correlation between the two variables • >0 -0,25 : Very weak correlation • >0,25 -0,5 : Sufficient correlation • >0,5 -0,75 : Strong correlation • >0,75 -0,99 : Very strong correlation • 1 : Perfect correlation Based on table 1.7, the correlation value is 0.600 which shows the magnitude of the relationship from variable X to Y1. This value if squared will produce a coefficient of determination (R square), namely 0.360. The coefficient of determination aims to determine the influence of the independent variable on the dependent variable. The result of the determination coefficient above is 0.360 which can be said that the X variable has an impact on Y1 by 36%. The simultaneous test table aims to determine whether there is a joint effect of the independent variable on the dependent variable. The existence of a joint effect between the independent variable and the dependent variable is shown if the calculated F value is greater than F in the table and the significance is less than 0.05. From table 1.8, the calculated F value is 47.163 and the significance is 0.000 so that the calculated F value is greater than F from the table (3.95) and the significance is less than 0.05, so it can be concluded that there is a simultaneous effect of variable X on Y1. Based on table 1.9, the variable X (pandemic) has a t value of 6.868 and a significance of 0.000 so that the significance value is less than 0.05 so that there is a partially significant impact of variable X on Y1 (trust). From the results of the unstandardized coefficient, the constant value is 13,493, which indicates that if the variable X is constant, the average value of the Y1 variable (trust) is 13,493. The regression coefficient of 0.686 shows that if the variable X increases by 1 unit, it will increase the Y1 variable by = 0.686. The (+) sign indicates that if the variable X increases, the Y variable will also increase. Based on table 5, the correlation value at 0.757 shows the magnitude of the joint relationship between variables X and Y1 to Y2. This value if squared will produce an R square value of 0.572. The coefficient of determination aims to determine the influence of the independent variable on the dependent variable. The result of the determination coefficient above is 0.575 which can be said that the variables X and Y1 together are able to influence Y2 by 57.5%. The simultaneous test aims to determine whether there is a joint impact of the independent variable on the dependent variable. Based on table 1.11 there is a joint impact between the independent variables if the calculated F value is greater than F in the table (3.95) and the significance is less than 0.05. The calculated F value is 55.573 and the significance is 0.000 so that the calculated F value is greater than F table and the significance is less than 0.05, it can be concluded that there is a simultaneous impact of variables X and Y1 on Y2.  Based on table 7. it can be seen that the variable X has a t-value of 2.084 and a significance of 0.040. A significance value less than 0.05 indicates a partially significant impact of variable X on Y2. The regression coefficient value of 0.205 indicates a positive impact, meaning that the higher the X value, the Y2 will increase. This value also means that Y2 will increase by 0.205 units for each one-unit increase from X. The variable Y1 has a value of t count 7.021 and a significance of 0.000. A significance value less than 0.05 indicates a partially significant impact of the Y1 variable on Y2. The regression coefficient value of 0.602 indicates a positive impact, meaning that the higher the Y1 value, the Y2 will increase. This value also means that Y2 will increase by 0.602 units for each one-unit increase from Y1.
Based on table 7 the impact between variables X and Y1, can be explained partially t-test statistics from variable X (pandemic) t-count result of 2.084, with t-

CONCLUSION
Based on the research result and discussion, it can be concluded that The direct impact of the Covid pandemic (X) on trust (Y1) is 0.205 (unstandardized) and 0.187 (standardized), the direct impact of trust (Y1) on the Y2 commitment is 0.602 (unstandardized) and 0.63 (standardized). Based on the t test, the t value (3.251) > t table (1.663), it means that there is an indirect impact between variable X (pandemic) on Y2 (commitment) through Y1 (trust). The impact of the Covid pandemic indirectly affects consumer commitment through the variable of trust with a combined direct and indirect impact of 1,099.
From these results, it can be concluded that the pandemic has a direct and indirect impact on trust and commitment. Managerial strategies that can be carried out by health care providers are the maximum use of information technology in socialization and services, including payment transaction services directed to non-cash, digital promotions, socialization , online health seminars / webinars, online consultations, digital technology homecare can be used for efficiency, reliability, and transparency. Improve social relations with consumers in addition to online with face-to-face with health protocols while prioritizing professional ethics of good health services, safety, friendliness to consumers, eliminating the stigma of fear/insecurity when it comes to health services including hospitals. Offline and online strategies can be done to strengthen good relationships with consumers in the long term so that consumers only trust the hospital's services.