Examining the treatment choice when getting a cold

Examine the relationship between socio-economic factors demographic, social capital, and risk attitude and treatment choice i.e: go to the doctor, self-medication, and non-drug treatment

Examining the Treatment Choice When Getting a Cold

By DANG HOANG HAI TRUONG

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

(SPECIALIZATION IN HEALTH ECONOMICS AND MANAGEMENT)

HO CHI MINH CITY, MAY 2015

Examining the Treatment Choice When Getting a Cold

A thesis submitted in partial fulfillment of the requirements for the degree of

MASTER OF ART IN DEVELOPMENT ECONOMICS

(SPECIALIZATION IN HEALTH ECONOMICS AND MANAGEMENT)

Major: Economics Code: 60310105

By DANG HOANG HAI TRUONG Academic Supervisor:

Dr NAM KHANH PHAM

HO CHI MINH CITY, MAY 2015

I guarantee data used in this thesis were truly collected through survey progress, along with using legal documents

The implications are withdrawn by the author from the experience of working, and learning

HoChiMinh city, 2nd May 2015

Dang, Truong Hoang Hai

Sometimes we need a convincer who knows better However it is difficult for us to judge whether his decision will lead to our expected outcome He also does not know what factors we are considering That is the reason why sometimes we make

an irrational decision which is from other’s point of view For a convincer, it is important to understand decision factors of person who is convinced In health aspects, doctor-patient communication is significantly vital For a health authority,

if he wants people to make a decision that benefit to the whole society, he also should understand their elements of consideration This thesis studies treatment decision making of patients when they get a cold Our survey shows that people have three kinds of decision when they get a cold i.e either “go to the doctor”, or

“self-medication” or “non-drug treatment” This thesis examines the impacts of psychological factors and socio-economic factors on each of the decisions The social pressure and patient’s perceived control factors do not have influence on the decision Females who have higher education are more likely to go to the doctor, and who have high income are less likely to choose self-medication We employed the multivariate probit model to analyze the treatment choice and a factor analysis

to construct psychological variables which were developed from the Theory of Planned Behavior

Khanh Nam, in spite of his busy schedule, he guided me to finish this dissertation Second, I would like to say my gratitude to my parents who encouraged me in hard moments and their unconditional love

Third, it would be a mistake if I did not make mention to my friends and my brother who helped me to sharpen the questionnaire

Finally, I express my gratitude to participants who play an important role in this thesis through their collaboration to complete the questionnaire

RUM Random Utility Maximization

TPB Theory of Planned Behavior

VND Vietnam dong

WHO World Health Organization

1.2 Research objectives……… ………… … 2

1.3 Scope of study……… 3

1.4 The structure of the thesis……… 3

CHAPTER 2: LITERATURE REVIEW……… ……….… 4

2.1 Key Concepts……… ………….… ……….………4

2.1.1 Common health problems and common cold…….………4

2.1.2 Patient choice and its special elements……….……… 4

2.1.3 Patient belief……… ……….4

2.1.4 Self-medication and economics of self-medication………… 5

2.1.5 Non-drug treatment…… ……… 6

2.1.6 Social capital… … ………6

2.2 Studies of socio-economic factors in health aspect……… ……… 7

2.3 Theory of Planned Behavior in Health Choice……… 10

2.4 Review of empirical studies ………11

2.5 Literature review conclusion……….12

CHAPTER 3: RESEARCH METHODOLOGY……… ……… …13

3.1 Analytical framework……….…13

3.2 Measurement of variables ……… 15

3.2.1 Qualitative process……… 15

3.2.2 Quantitative process……… ……… 18

3.2.2.1 Indirect measure of the Theory of Planned Behavior…….18

3.2.2.2 Direct measure of the Theory of Planned behavior 21

3.2.2.3 Socio-economic……… 22

3.3 Econometric Models……… 22

3.4 Variable description……… ……… 24

3.5 Research strategy ……… 29

3.5.4 Data analysis………31

3.5.5 Data framework… ……… 31

CHAPTER 4: RESEARCH RESULTS……… ……… ….… ……… 32

4.1 Overview of Vietnamese health environment……… …… 32

4.2 Descriptive statistics……….………… 32

4.2.1 Psychological factors statistic……… …………33

4.2.1.1 Attitude……… ……….33

4.2.1.2 Subjective norm……….36

4.2.1.3 Perceived behavioral control……… ………37

4.2.2 Socio-economic statistic……… 38

4.2.2.1 Demographic variables……… 38

4.2.2.2 Descriptive social capital variables………40

4.2.2.3 Descriptive risk variables……… 41

4.3 Regression results……….……… 42

4.3.1 Theory of Planned Behavior……… ……… 42

4.3.1.1 Indirect measure……….42

4.3.1.2 Direct measure……… 49

4.3.2 Socio-economic factors……… ……… 51

CHAPTER 5: CONCLUSIONS AND POLICY IMPLICATIONS…… 55

5.1 Conclusions…… ……… ……… …55

5.2 Policy Implications……… ……….…… …… 56

5.3 Other suggestions…….………….……….…… ……… 57

5.4 Limitation……… ……… ………… 58

REFERENCES.……….… ………… 59

APPENDIX A……… ……… 64

APPENDIX B……… ……….….……71

Figure 2 Data framework……… ……….31

Figure 3 The income and the choice……… … 39

Figure 4 The gender and the choice……… ……….40

Figure 5 The social group adherence and the choice……… …42

LIST OF TABLES Table 1 Theories of socio-economic in health aspect……….……….9

Table 2 Salient belief items……….18

Table 3 Variables definition……….……… 25

Table 4 The statistical results of the choice……….………33

Table 5 The attitude and the choice……….……… ……… 35

Table 6 The subjective norm and the choice……… ……… 37

Table 7 The perceived behavioral control and the choice………38

Table 8 Descriptive demographic variables……….38

Table 9 Joint social group……….……… 40

Table 10 Description of trust……….……… 41

Table 11 Risk attitude description……….……… 41

Table 12 Factor analyses and Cronbach’s alpha of the indirect measure…….… 43

Table 13 Multivariate probit regression of indirect measure……….… 47

Table 14 Factor analyses and Cronbach’s alpha of the direct measure………… 49

Table 15 Multivariate probit regression of direct measure……….………50

Table 16 Factor analyses and Cronbach’s alpha of the socio-economic………….51

Table 17 Multivariate probit regression of socio-economic……….……… 52

a result of lacking of patients

The ability of “downstream” hospitals, by doubting not only about technique ability but also service ability could damage deeply in patient’s belief Even when they can completely cure the common diseases, people still want to stay away and move to higher level hospitals

There is information asymmetry between the patients and medical staffs For instance, patients do not know whether drugs they are using are good or bad for their health and also do not know which are the good sides and which are the drawbacks of the medicine services that they are consuming as well as the opportunity cost they have to pay when they could get to other hospitals or choose to not experience in medical care Moreover, until now in medicine area, for a certain health problem, we still do not have powerful measure tools to predict which doctor, or drug, is better and completely guarentee for a positive outcome There are chances existing, chances for getting cured, and chances for complication The result is mostly on individual basis, which one experiences himself as a patient or who know the patient, don’t know thoroughly about the situation In this lack of information and uncertainty environment, individual seems

to make a choice between various solutions by their feeling and belief

Health is a vital matter for each individual Good health allows us to live and work more efficiently and positively Bad health on other hand could negatively affect people attitude and productivity That is the reason why an increase in level of citizens’ health can improve the productivity, GDP, research, or education so that the social community can be enhanced in all areas In the other side of the coin, when a health problem

appears, it is important to find a suitable method which can cure the disease with appropriate cost, not only be measured by money, but also by time, career, health, relationship, etc

This thesis stays focused on in aspect how health’s belief impacts on health making In order to explain this relationship, common cold is used as a common health problem Common cold’s nature is a popular health problem and patients have many choices

decision-The author believes that investigation of this event could lead to a new insight, especially in Vietnam, a country that lacks of health behavior research Moreover, successful explanation of this relationship provides a suggestion on communication with patient Since talking with doctor takes a main part in making patients satisfy (Bensing 1991; Maguire & Pitceathly 2002), and health decision-making is based on belief, this thesis will become a milestone for healthcare practice

2 Examine the relationship between psychological factors and treatment choice, i.e go

to the doctor, self-medication, and non-drug treatment

3 Examine the relationship between socio-economic factors (demographic, social capital, and risk attitude) and treatment choice i.e: go to the doctor, self-medication, and non-drug treatment

1.3 Scope of study:

This study compares the causes of treatment decision-making: go to the doctor, medication, and non-drug treatment To be specific, the subjective of research is common-cold, a common health’s problem that patients have many solutions Data are received by survey method The location and time to survey is district 3, from April to May, 2015 Collected data have been analyzed by multivariate probit model

self-Data have been collected by cluster sampling method District 3 citizens who are at ages from 18 to 80 are selected

1.4 Structure of the thesis:

The thesis is divided into five chapters Chapter I is the introduction that presents the research problem, as well as research objectives and scope of study Chapter II is the literature review which defines concepts, provides theoretical background, and discusses empirical related studies Chapter III is the methodology which gives the method of study, including analytical framework, measurement variables, qualitative progress and variable definition After data are analyzed, the result of study appears at chapter IV Chapter V is the conclusions and political implications

CHAPTER 2: LITERATURE REVIEW 2.1 Key Concepts:

2.1.1 Common health problems and common cold:

Everyone wants to live a healthy life, and want to recover from disease However, there

is usually not only one option to solve the problem Specially, in common health problems, people have more solutions than in serious health events

Common cold is a common health problem mostly because of virut From patient’s point of view, Mayo Clinic’s definition is suitable According to Mayo Clinic, healthy adults are expected to have a few cold each year, and most people recover from a common cold in about a one or two weeks Sign and symptoms include runny or stuffy nose, cough, and mild fatigue Not all cases need to see a doctor and these patients only seek medical attention when getting high temperature fever Fever is accompanied by sweating, chills, and a cough with colored phlegm, severe sinus pain

2.1.2 Patient choice and its special elements:

Patient choice as Community Pharmacy Standards of Practice declared in 2015, relates

to rights, which is that no one can enter into agreement with any person, who limit a patient’s choice of pharmacy, excepts as required or permitted under the bylaws

Dixon et al (2010) gave evidence that the patient valued the ability to make choice They also had evidences that patient rely on their own or others’ experience to inform their choice

There are different points of patient and doctor when comparing to normal buyers and firms Dixon et al (2010) also argued that patients do not act as normal consumers, they cannot test the product before consuming or even after having consumed it It is also hard for patients to find out relevant information to their condition, and they place trust

on the provider Externalities are also an important factor in health aspect

2.1.3 Patient belief:

As mentioned before, patients have to place their trust on the providers Therefore, the concept of patient belief is necessary to join the thesis

Groot and Steg argued that general belief may affect a wide range of behaviors (Groot and Steg, 2007:3), that means we need to explore belief to research behavior and health decision-making

Mostly because of the asymmetric information, patients choose health solutions mainly based on their belief Because belief is the psychological state in which an individual holds a proposition or premise to be true ( Passos et al., 2013:1) Knowing each patient’s health beliefs is the key feature of the new doctor- patient encounter This may lead to a negotiated treatment plan to which both patient and doctor can adhere (Vermiere et al., 2001:400) This empirical research impresses the important role of understanding patient’s belief to help them improve their health behavior

There are a lot of theories that study the relationship between belief and decision Among them is the Theory of Planned Behavior (TPB) Moreover, there have been a lot

of meta-analyses and reviews of TPB In Armitage work in 2001, he concluded “TPB is

a useful model for predicting a wide range of behaviors and behavioral intentions.” (Armitage and Conner, 2001:340) or Armitage’s another argued in the next two years,

“At present, the theory of planned behavior is arguably the most dominant model of attitude-behavior relations” (Armitage and Christain 2003:5)

2.1.4 Self-medication and economics of self-medication:

This thesis combines psychological factors and socio-economics factors to explain treatment decision-makings: go to the doctor, self-medication or non-drug treatment

In a study in Jordan about medication (Yousef et al.,2007:24), Yousef said medication is a concept which belongs to self-care and can be defined as consuming medicines without the advice of physician for diagnostic, or treatment As the result of World Self-Medication Industry, Self-medication is popular in both developed and developing countries

self-Self-medication is an important matter in healthcare aspect It also was defined as the consuming of patient without physician’s advice or on the advice of pharmacist or on his own decision (Yousef et al., 2007:24) Patients usually use self-medication to solve

their normal health problems World Self-Medication Industry also declared: medication is the treatment of common health problems with medicines” On the other hand, “Most of the self-medication was involved with headache and fever, cough and cold” (Verma, 2010:60), which related to the sign and syndrome of common cold The reason why patients choose to use self-medication method is its benefits Self-medication users save their time and money, compared to professional care users (Chang et al., 2003:721) However, self-medication users increase risks of drug induced and/or drug-resistant strains (Chang et al., 2003:721)

“Self-There are many studies that discovered the link between self-medication and economic (Afolabi, 2012) Among them is the study of Chang et al (2003) They built

socio-a RUM model of self-medicsocio-ating behsocio-avior of consumers who bsocio-alsocio-ance the socio-advsocio-antsocio-ages and disadvantages of self-medication Their theory formulates the connection between income, health insurance and self-medication decision

2.1.5 Non-drug treatment:

WHO (1994) argued not all health problems need drug to recover Along with drug treatment, non-drug treatment is used in many health problems: Pain control, hypertension, Acute Diarrhea with mild dehydration in child, open wound, etc (Yurdanur 2012; Institute for Quality and Efficiency in Health Care 2011; WHO 1994; Medical Protection Society 2012) and promised many benefits For common cold treatment, non-drug treatment also plays an important role It is the first line treatment (Lousianaphamacists update)

Furthermore, non-drug treatment has many strong points such as reducing number of drug misusers, drug abusers, waiting time, improving access to medicine for necessary patients

2.1.6 Social capital

Putnam (2001) in his article Social Capital defined that social capital is networks and the associated norms of reciprocity that have value

Pherson et al (2013) found out that family and community social capital are important

in health aspect Another empirical research came from Pampel, F et al (2010) They declared that socio-economics can affect the incentives or motivations for healthy behavior

Social factor also plays an important part of socio-economic theories (Svendsen and Sorensen, 2006) A review of socio-economic theories and the components are necessary because this thesis’s objective is to examine the relationship between socio-economic and treatment decision-making

2.2 Studies of socio-economic factor in health aspect:

Social capital:

Rocco and Suhrcke’s (2012) research declared three mechanisms of social capital for the determinants of individual health: the increasing in accessing to relevant health’s information, the providing of informal healthcare and support in health problems, and finally, well-organized, connected groups are more effective in lobbying in the political economy mechanism In additional, Grootaert et al., (2004) argued that social capital should include the trust in people, and the adherence to any social group

Demographic characteristics:

Belief is a core aspect to explain behavior However, kinds of patient also affect the selection and therefore health status For example, patients with high education demonstrated in Korsh’s work are more likely to express their fears and hopes to have a better chance (Korsh et al., 1968:14) Similar to Korsh, Mocan and Altindag pointed out the relationship between health and education “this positive relationship between education and health is robust whether one analyzes aggregates (e.g., mortality or morbidity rates ) or micro units (e.g., individual’s self- reported health status, or sick days)” ( Mocan and Atindag, 2013:1) Another research that came from Korean, Shin and Kang explored health behaviors, and examined health behaviors in relation to demographic factors (Shin and Kang, 2014:1)

Risk attitude:

There is a connection between risk perceived and behavior, as well as decision making Health belief model, which was created by U.S Public Health Service in 1950s, indicates that when a person perceives chances of risk, they are more likely to prevent it from happening Therefore risk attitude has a role to the behavior Szrek et al (2012) predicted actual risk behavior in the domain of health They argued that different persons perceive different risks

Rormann (2002) also believes that people are different in risk attitude

Therefore the diversity of people’s behaviors and decision making depends on their risk- taking propensity

Table 1 Theories of socio-economic in health aspect:

socio-economic used

Is Social Capital Good

for your Health?

Rocco and Suhrcke

and Decision Making

Szrek et al (2012) Risk attitude

Risk Attitude Scales:

Concepts and

Questionnaires

Rormann (2002) Risk attitude

Source: Thesis’s Author

2.3 Theory of planned behavior (TPB) in Health Choice:

To understand clearly patient belief, as well as to measure psychological factors and examine its impact on treatment decision making, this dissertation is based on the theory of planned behavior, a theory about the connection between belief and behavior (Ajzen, 1991)

A lot of empirical studies have been using TPB to discover the behavior, such as Knowlden et al (2012) predicted the sleep intentions and behaviors, Ziadat (2014) applied TPB in Jordanian Tourism, and Rhodes (2008) learned the behavioral physical activity

TPB is also effective in prediction and intervention of health behavior, as Darker announced “This extended TPB has been generally supported as a framework for developing and testing such intervention [health intervention]” (Darker et al., 2009:16) Other academics adopt a similar position, such as Baban and Cracium in their review

“it [TPB] states how these constructs should be operationalized, which makes the design of behavior change interventions easier” (Baban and Craciun, 2007: 5)

TPB is a standard theory about belief that a lot of studies are using to explain behavior, Groot and Steg declared that studies based on the TPB scarely examined more general behavioral determinants, such as values or general beliefs (Groot and Steg, 2007:3) Cognitive structure is the collection of beliefs that individuals consider an objective and how these beliefs are constructed in memory ( Alba and Hutchinson, 1987)

Belief: Belief could be recognized as the information that people have about behavior: its likely outcome, the normative judgment of others, and the likely ability to its performance (Ajzen, 1991 )

Ajzen and Fishbein (1991) created a great theory to determine elements of people’s action based on belief, (named theory of planned behavior) According to them, behavior is a result of three factors

The first is belief about the likely outcome of the behavior (behavioral belief) and the evaluations of these outcomes, that both create the attitude

The second is belief about the normative expectations of others (normative belief) and the motivation to comply with these expectations, these two factors combined to become subjective norm

The final one is belief about the present of factors that may facilitate or impede performance of the behavior (control belief) together with belief about the perceived power of these factors, to become perceived behavioral control

The attitude, the subjective norm and the perceived behavioral control are three forces for people’s behavioral intention, and they lead to actual behavior

Moreover, unlike other factors, perceived behavioral control plays an important role to the actual behavior

2.4 Review of empirical studies:

Theory of planned behavior shows out the method to predict and intervene human behavior (Ajzen,1991) in many areas, such as healthcare (Arbour-Nicitopoulos et al, 2009), teaching (Jeong and Block, 2011), risk (Turchik and Gidycz, 2012)

For the predicting health behavior purpose, Gerend and Shepherd (2012) use both the theory of planned behavior and the health belief model to predict human papillomavirus vaccine uptake The result was that the theory of planned behavior outperforms the health belief model in the purpose of predicting the behavior Another research predicting the TPB firstly concern about the attitude Aladjem (2010) via two essays with 135 individuals in essay 1and 131 individuals in essay 2 had said out the connection between attitude and the product involvement on judgment and choice The second element of TPB is the norm We can find it out from studies that use extended technology acceptance model in testing the impact of between subjective norm and e mail usage (Mutlu and Ergeneli, 2012), or theory of planned behavior to determine the contribution of moral norm on behavior of HPV vaccination (Ilona et al., 2011)

And last but not least, perceived behavior belief is a component that makes TPB special Kraft et al (2005) operated a study that exploring the construction of perceived behavioral control They discovered the important role of controllability in predicting behavior This research also suggested that measuring perceived behavioral control by means of perceived difficulty

Martin et al (2010) discovered the impact of perceived behavioral control on the control of gambling behavior Research discovered that college students are more likely

to be vulnerable to gambling problem Authors implicate that to decrease gambling activity of college students, increasing student’s perception of their ability to control gambling is necessary

Chang et al (2003) used World Bank’s Living Standards Measurement Survey of Vietnam’s data, 1997-1998 to develop self-medicating behavior’s model They discovered that self-medication is an inferior [good] at high incomes levels and a normal [good] at low income levels In the other hand, besides the young male group, the data’s result showed up that there is a negative impact of income on pharmacy visit

2.5 Literature review conclusion:

This dissertation uses the psychological and socio-economic approach to examine the treatment choice, using common cold as common health problem to investigate patient’s belief and their treatment decision In order to reach the aim, this dissertation’s framework mainly applies the TPB, a theory which successes in explaining wide range

of behaviors According to the theory, human behavior is guided by three kinds of beliefs: beliefs about the likely consequences of the behavior, beliefs about the normative expectations of others, and beliefs about the presence of factors that may facilitate or impede performance of the behavior On the other hand, health behavior and health status are also controlled by demographic factors

Socio-economic is another main component for treatment choice Its empirical studies suggest that there is a connection between income factor and treatment choice

CHAPTER 3: RESEARCH METHODOLOGY

3.1 Analytical framework

Figure 1 Model of the study PSYCHOLOGY

Attitude Subjective Norm

Perceived Behavioral

Control

Source: Thesis’s Author

TREATMENT CHOICE

Theory of planned behavior:

Perceived behavioral control:

Perceived behavioral control is analyzed as how easy or difficult patient evaluate, and it

is depended on patient’s experience as well as their prediction about the obstacle toward the behavior

This research analyzes the patient’s perceived behavioral control by investing if they have enough time, or money, or can be able to access to treatment method, or they would perceive control over the treatment period

Socio-economic factors:

Socio-economic includes demographic characteristic, social capital and risk behavior The demographic characteristics consist of gender, age, education, married, family, and income

Social capital: social capital is measured by three factors, did or did not attend in a

social group one year ago, the trust in people surrounding, and the trust in people from

different careers

Risk attitude: Trend to take risk

Treatment choice:

Patient’s treatment choices are their solutions when getting a cold They are “go to the

doctor”, self-medication, and non-drug treatment

Patient’s choice is analyzed by asking them choose their treatment method

Psychological factors and socio-economic factors are two components that patients

consider to make choice when they get a cold

3.2 Measurement of variables:

3.2.1 Qualitative process:

According to Ajzen (2010), TPB variables could be calculate through direct or indirect

measure In order to collect material for indirect measurement, the author asks for

salient beliefs Model salient beliefs can provide the basis for constructing a standard

question (Ajzen, 2010) Salient beliefs when coping with common cold problem have

been elicited to salient behavioral outcome, salient normative references, and salient

control factors

Before asking for the salient beliefs, it is necessary to ask their choice when they get a

cold

Salient behavioral outcomes:

a What do you believe are the advantages of your choice when you get a cold?

b What do you believe are the disadvantages of your choice when you get a cold?

Salient normative beliefs:

a Are there any individuals or groups who would approve of your choice when you get

Salient control beliefs:

a What factors or circumstances would enable you to do your choice when you get a cold?

b What factors or circumstances would make it difficult or impossible for you to do your choice when you get a cold?

Results of the pilot study:

The pilot result has been classified Then the author has withdrawn the most chosen items

Generally, the participants have 3 choices: go to the doctor, self-medication, non-drug medication

The pilot study collected 21 samples Among them average sickness times per year is 3,

in range from 1 to 10 Two-thirds are married Their old age is from 21 to 69, with mean value are 31 Almost have insurance Nearly a half of them choose change life style without using drug solution One-fifth wants to go to the doctor Their referents focus on family, friends, co-workers and medical staffs The elements that they mainly consider are fast and slow recovery, drug side effect, and time saving Almost dislike

using drug, however they prefer keeping their productivity That is the main reason why they choose self-medication solution

The belief items which elicited from pilot study:

a Salient Behavioral outcomes depend on the participant’s choice:

A Go to the doctor: There are 3 belief items:

Inconvenience, Much money costing, and Time costing

B Self-medication: There are 2 belief items:

Convenience, and Time saving

C Non-drug treatment: There are 4 items:

Better life style, Less medication taking, Improving immune system, and Rising body’s protection ability

b Salient Normative beliefs result: There are 4 items:

Family, Friends, Co-workers and Medical staffs

c Salient Control beliefs: There are 2 items:

Inconvenience, and Disease more serious

Table 2 Salient belief items

Choice of Self-medication Time saving

Better life style

Choice of Non-drug treatment

Less medication taking Improving immune system Rising body’s protection

ability

Salient normative

beliefs

Family Friends Co-workers Medical staffs

Salient Control

beliefs

Inconvenience Disease more serious

Source: Thesis’s Author

3.2.2 Quantitative process:

3.2.2.1 Indirect measure of the Theory of Planned Behavior:

Based on qualitative process’s result and the guiding of Ajzen (2010), the author of this thesis has built elements of indirect measures They are Behavioral beliefs, Normative beliefs, Control beliefs Moreover, before releasing the questions, the author of this thesis has asked directly to people surrounding to find out the word which suit to Vietnamese culture

Measure Behavioral beliefs:

To measure behavioral beliefs, two questions for each behavioral belief items are necessary to ask One is about Behavioral belief strength (b), and one is about Outcome evaluation (e)

Behavioral belief strengths of “go to the doctor” choice are the patient perceptions of the outcome of “go to the doctor”: Inconvenience, Much money costing, and Time costing

Outcome evaluations of “go to the doctor” choice are: Inconvenience is dissatisfying, Much money costing is not necessary, and Time costing is necessary

Behavioral belief strengths of self-medication choice are Convenience, and Time saving

Outcome evaluations of self-medication choice are: Convenience is satisfying, and Time saving is necessary

Behavioral belief strengths of non-drug treatment choice are Better life style, Taking less medication taking, Improving immune system, and Rising body’s protection ability

Outcome evaluations of non-drug treatment choice are: Better life style is helpful, Using less medicine is helpful, Improving immune system is helpful, and Rising body protection ability is helpful

They are all measured by bipolar disagree-agree scale, scored -3 to +3

The result then collected and Behavioral belief strength (b) is multiplied by Outcome evaluation (e) of each type of attitude Attitude toward treatment decision making’s components are ATTAktt, ATTAmnt, ATTAmtg, ATTBtt, ATTBtg, ATTCls, ATTCut, ATTCtbv, ATTCdk Factor analysis will be used to get factors of Attitude toward treatment decision making

Measure normative beliefs:

According to Ajzen’s guild, to measure Normative beliefs, three questions for each referent are necessary to ask These questions consist of Injunctive norm (i),

Descriptive norm (d), and Motivation to comply (m) They are measured by bipolar disagree-agree scale, scored -3 to +3

Injunctive norms are: Your family approves your choice, Your friends approve your choice, Your co-workers approve your choice, and Medical staffs approve your choice Descriptive norms are: Your family performs like your choice, Your friends perform like your choice, Your co-workers perform like your choice, and Medical staffs perform like your choice

However Motivation to comply variables are measured by unipolar unlikely- likely, scored 1 to 7

Groups that are the subjective of Motivation to comply: Family, Friends, Co-workers, and Medical staffs

Injunctive norm (i)

Then Injunctive norm (i) is multiplied by Motivation to comply (m) and Descriptive norm (d) is multiplied by Motivation to comply (m) to become Normative beliefs They are SN1gd, SN1bb, SN1dn, SN1yt, SN2gd, SN2bb, SN2dn, SN2yt Then the author has used factor analysis to get Subjective norm’s factors

Measure Control beliefs:

According to Ajzen, to measure Control beliefs, two questions for each item are necessary to ask One for Control belief strength (c), one for Power of control factor (p)

Control belief strength items are: I think it is not convenient to perform my choice, and

I think perform my choice may lead to more serious outcome They are measured by bipolar disagree-agree scale, scored -3 to +3:

Power of control factor items are: Not convenience, Disease more serious They are measured by bipolar difficult-easy scale, scored -3 to +3

Control belief strength (c) is multiplied by power of control factor (p) to become Control beliefs They are PBC1ktt, and PBC1bnh Then the author has used factor analysis to get Perceived behavioral control’s factor

3.2.2.2 Direct measure of the Theory of Planned Behavior:

Direct variables include Attitude toward behavior, Subjective norm, Perceived behavioral control

Measure Attitude toward behavior:

According to Ajzen’s guild (2010), Attitude toward behavior could be measure by asking 2 questions They are Experiential Attitude and Instrumental Attitude They are measured by bipolar bad-good scale, unsatisfied-satisfied, scored -3 to +3:

When you get a cold, with your choice to recover, what do you feel?

Bad: -3 : -2 : -1 : 0 : 1 : 2 : 3 : Good

Unsatisfied: -3 : -2 : -1 : 0 : 1 : 2 : 3 : Satisfied

Factor analysis is used to calculate attitude toward behavior

Measure Subjective norm:

According to Ajzen (2010), Subjective norm could be measure by asking two statements They are Injunctive norm and Descriptive norm They are measured by bipolar disagree-agree scale, scored -3 to +3:

Most people approve me to do my choice when I get a cold:

Disagree: -3 : -2 : -1 : 0 : 1 : 2 : 3 : Agree

Most people will do as my choice when they get a cold:

Disagree: -3 : -2 : -1 : 0 : 1 : 2 : 3 : Agree

Factor analysis is used to calculate Subjective norm

Measure Perceived behavioral control:

According to Ajzen (2010), Perceived behavioral control could be measured by asking:

When I get a cold, I am confident that I can perform my choice:

by scale from 1 to 5, anchored by completely disagree and completely agree

Risk attitude is measured mainly based on the participant’s trend of risk taking

Y1: Go to the doctor Y1= 0: not choose “go to the doctor”

Y1=1: choose “go to the doctor”

Y2: self-medication Y2= 0: not choose self-medication

Y2=1: choose self-medication

Y3: non-drug treatment Y3= 0: not choose non-drug treatment

Y3=1: choose non-drug treatment

X: Vector of psychological variables, i=1: Attitude toward Behavior

i=2: Subjective Norm i=3: Perceived Behavioral Control Z: Vector of socio-economic variables, j=1: Demographic characteristics

j=2: Social capital j=3: Risk attitude

Multivariate probit model is used for research objective 2 and 3: to examine the relationship between psychological factors and treatment choice; socio-economic and treatment choice

Multivariate probit model belongs to models for discrete choice Multivariate probit model allows discover the correlation between unobserved factors and different choices Patients can choose different solutions at the same time The model uses maximum likelihood estimation According to Cappellari and Jenkins (2006:157),

“maximum likelihood estimation works by simulating likelihoods and then average those these” As Greene (2003) mentioned, the log of likelihood function is:

ln L(Ɵ|y)=∑𝑛 ln⁡𝑓

𝑖=1 (yi|Ɵ)

The logic of the maximum likelihood is easily illustrated in the setting of a discrete distribution Consider a random sample of the following 10 observations from a Possion distribution: 5, 0, 1, 1, 0, 3, 2, 3, 4, and 1

Consider maximizing ln L(Ɵ|y) with respect to Ɵ Since the log function is monotonically increasing and easy to work with, we usually maximize ln L(Ɵ|y) instead: In sampling from a Poisson population,

ln L(Ɵ|y)= -nƟ+ lnƟ∑𝑛𝑖=1 yi- ∑𝑛𝑖=1𝑙𝑛(yi),

ǝln⁡L(Ɵ|y)

ǝƟ = -n+1

Ɵ∑𝑛 𝑖=1 yi=0 => ƟML=⁡𝑦̅n

3.4 Variables description:

Demographic data and variables are coded Furthermore, information received from the survey will be classified to each group which represent to model variables (See Table 3)

Table 3 Variables definition Elements of

the Model

Thesis

Questionnaire Item Number

to the doctor choice

Patient’s attitude about the salient belief of time costing, toward go

to the doctor choice

Patient’s attitude about the salient belief of much money costing, toward go to the doctor choice

Patient’s attitude about the salient belief of convenience, toward self-medication choice

Patient’s attitude about the salient belief of time saving, toward self-medication choice

Patient’s attitude about the salient belief of better life style, toward non-drug treatment choice

Patient’s attitude about the salient belief of using less medicine, toward non-drug treatment choice

Patient’s attitude about the salient belief of improving immune system, toward non-drug treatment choice

Patient’s attitude about the salient belief of rising body’s protection ability, toward non-drug treatment choice

Injunctive normative belief of patient’s friends (the strength of the friend’s approve or not approve the choice multiplied by the motivation to comply)

Injunctive normative belief of

4B, 4C, 4D, 4E,

4G, 4H, 4I,4J, 5B, 5C, 5D, 5E

patient’s co-workers (the strength

of the co-worker’s approve or not approve the choice multiplied by the motivation to comply)

Injunctive normative belief of medical staffs who related to patient (the strength of the medical staff’s approve or not approve the choice multiplied by the motivation to comply)

Descriptive normative belief of patient’s family (the strength of the belief that family will or not do the same when getting a cold multiplied by the identification)

Descriptive normative belief of patient’s friend (the strength of the belief that friends will or not do the same when getting a cold multiplied by the identification)

Descriptive normative belief of patient’s co-worker (the strength

of the belief that co-workers will

or not do the same when getting a cold multiplied by the identification)

Descriptive normative belief of medical staffs who related to patient (the strength of the belief that medical staffs will or not do the same when getting a cold multiplied by the identification)

Salient belief control factor about the more serious disease of the treatment choice

Instrumental in nature component

of the overall evaluation of the treatment choice

Experiential quality component of the overall evaluation of

2

3

measure) thetreatment choice

Surrounding’s injunctive quality

of the treatment decision (About the treatment approval or disapproval of most people)

Surrounding’s descriptive quality

of the treatment decision (whether most people choose the same treatment decision)

6

Demographic

characteristic

Age Gender

Married

Fami income

edu

The age of the participant

Gender of participant

0: female 1: male Marital status

0: not yet married

1: married

Number of people who living in same house

Income per month

0: lower or equal to 10 million Vietnam dong incomes per month 1: higher than 10 million Vietnam dong incomes per month

Education level of participant

0: lower than high school 1: high school or higher

Social capital

Codetgtc

kvt ahli dygd kcmt bth

Belongs or not to a social group

0: not attend to any social group

1: attend to at least a social group

Every one surrounding is trustable

Nobody surrounding harm you because of their benefit

Every one surrounding helps you when you need

It is not wise to lend people surrounding

I trust shopkeepers

I trust local government officers

I trust central government officers

9 9a 9b 10A, 10B, 10C, 10D 11A,

dp

tu

cs

gv ddbs

11D, 11E, 11F, 11G

Go to the doctor Self-medication Non-drug treatment

1a 1b 1c

Source: Thesis’s Author

3.5 Research strategy

The goal of this research is to “identify what kind of believe the patient have when they have common health problem”, and to “explore the element that take the most important place in health decision of each type of patient” The research strategy is a descriptive correlational study combined with cross- sectional research survey Multivariate probit model is used This strategy is suitable to reach the research objectives Descriptive study describes the concept belief in health aspect Correlational study determines the relationships between beliefs and traits in socio-economic to health behavior in common health problem situation Multivariate probit model is choice to explore the quantitative relationship between variables

The research takes two steps Firstly, this thesis approaches to TPB by qualitative method 20 persons were deep interviewed The questionnaire is based on Ajzen’s TPB Questionnaire and Construction (2010) and Glanz et al (2008), although it was modified for adapting to Vietnamese condition The result of this pilot test was used to develop measurement scales for TPB variables

Secondly, in quantitative process, this thesis builds standard questionnaire bases on qualitative result and socio-economic variables

Both questionnaires are built in Vietnamese Before using the questionnaire to collect data, the author has also run a deep interview to test the friendly character of the questions, and modified it to better version These affords aim to make it suitable to survey the participants Then it is translated to English to fit with this thesis

Multivariate probit model is used to analyze the relations between variables and to compare between different choices

3.5.1 Setting:

Participants come from in district 3, HoChiMinh city, VietNam District 3 is located in the middle of HoChiMinh city, and contains 14 wards It is one the most active area with the top living standard District 3 was chosen because it has diversity population

It’s high living standard attracting lots of people coming to live and work The diversity

of population is necessary to get information for this empirical study

3.5.2 Sampling technique and sample size:

This study selected 14 wards in district 3 Nearly 12 households were interviewed at each ward The list of household selected was created randomly This cluster sampling

is used because this empirical study needs a method that is cost effective and time saving Comparing to simple random technique, cluster sampling is less precious To reduce the disadvantage, this thesis uses a plenty of clusters, with small cluster size Ahmed et al (2009) in published “Methods in Sample Surveys” has recommended using as many of cluster as feasible, small cluster size to have increase cluster method’s effect

3.5.3 Data collection process:

After analyzing pilot study to complete the questionnaire, it is used as trial to test if it is easy to approach for people The questionnaire has to be modified to keep its own nature and let people understand clearly

Two groups are organized to operate the survey There are two members in each group One member asks and explains while the other notes the answer in the questionnaire They are instructed to understand the idea, main theory contain and the meaning of each question When they conduct the survey, they take note the participant’s ideas that they don’t clearly understand These notes will be returned to the author with the answer to confirm the result

When they collect enough survey, they return them to the author The author checks if they are all qualified The unqualified survey will be identified the address and error points It is necessary to conduct another survey at the same person

Data of the survey will be entered Microsoft excel to prepare for Stata statistical analysis

3.5.4 Data analysis:

Descriptive statistic is used to have a general look at the thesis’s data

Factor analysis is used to combine variables of the same group The result is used as a variable which represent value of that group

Multivariate probit model is used to find out the relationship between 3 choices and their causes

3.5.5 Data framework:

Figure 2 Data framework

Source: Thesis’s Author