**2.3.2 HRQOL measures applicable to CHB**

Health-related quality of life (HRQOL) can be measured by generic and disease specific measures (M.S. Bayliss, 1999; Brown, 1999; Eisen et al., 1999; J. J. Gutteling et al., 2007; Yacavone et al., 2001; Younossi, 2001; Younossi & Guyatt, 1998). Generic HRQOL measures can be applied to different patient populations (M.S. Bayliss, 1999; Eisen et al., 1999; J. J. Gutteling et al., 2007; Yacavone et al., 2001; Younossi, 2001; Younossi & Guyatt, 1998). The advantage of using this instrument is that it can compare with other types of diseases or healthy control population (M.S. Bayliss, 1999; Eisen et al., 1999; J. J. Gutteling et al., 2007; Yacavone et al., 2001; Younossi, 2001; Younossi & Guyatt, 1998). Therefore, it is widely used in health services and comparative studies. However, generic measures may not detect small but important clinical changes specific to a particular patient group. The Medical Outcomes Study Short Form-36 (SF-36) is the most commonly used (J. J. Gutteling et al., 2007; Yacavone et al., 2001; Younossi, 2001; Younossi & Guyatt, 1998). It showed in a study by Foster et al. that CHB patients had significant lower HRQOL scores in mental health and general health perception aspects, but their physical related HRQOL scores were comparable to the healthy control (Foster et al., 1998). Previous studies on the use of SF-36 on CHB patients have demonstrated that patients with less severe disease had lower HRQOL scores in general health compared to those with general population or healthy controls (Bondini et al., 2007; Foster et al., 1998; S C Ong et al., 2008; Tan et al., 2008). Once they developed complications, lower HRQOL scores was found in both physical and mental health (S C Ong et al., 2008). The effectiveness of anti-viral treatment was detected by the SF-36 in a longitudinal study on 150 Chinese CHB patients at different stages of illness receiving LVD treatment (Yi, 2006).

A disease-specific measure theoretically can detect small but clinically important changes on HRQOL that are unique to the particular condition although it does not allow for comparison with the general population or other disease groups (M.S. Bayliss, 1999; Eisen et al., 1999; Younossi, 2001; J. J. Gutteling et al., 2007; Yacavone et al., 2001). Several HRQOL measures specific for chronic liver disease patients, such as the Chronic Liver Disease Questionnaire (CLDQ) (Younossi, Guyatt, Kiwi, Boparai, & King, 1999), the Hepatitis Quality of Life Questionnaire (HQLQ) (M. S. Bayliss et al., 1998), the Liver Disease Quality of Life Questionnaire (LDQOL) (Gralnek et al., 2000), the Liver Disease Symptom Index (LDSI 1.0 and 2.0) (Unal et al., 2001; van der Plas et al., 2004), the Hepatitis B Quality of Life (HBQOL) (Spiegel et al., 2007) and Chronic Liver Disease-Specific Quality of Life (CLD-QOL) (Lee et al., 2007) are available. Each instrument has its

al., 2001; Sobhonslidsuk et al., 2006). Data from Hussain et al found there was weak correlation between level of education and physical HRQOL (Hussain et al., 2001). On the other hand, another study proved that patients with lower education level had significant

Chronic co-morbidity also affected HRQOL in patients with chronic liver disease (Hauser, Holtmann, et al., 2004; Hauser, Zimmer, et al., 2004; Hussain et al., 2001). Hauser et al examined 94 patients with CHC attending a liver clinic and showed that psychiatric comorbidities was one of important determinant of mental component summary (MCS) score of SF-36 (Hauser, Zimmer, et al., 2004). The number of active co-morbidities was associated with the SF-36 physical component summary (PCS) score (Hauser, Zimmer, et al., 2004).

Health-related quality of life (HRQOL) can be measured by generic and disease specific measures (M.S. Bayliss, 1999; Brown, 1999; Eisen et al., 1999; J. J. Gutteling et al., 2007; Yacavone et al., 2001; Younossi, 2001; Younossi & Guyatt, 1998). Generic HRQOL measures can be applied to different patient populations (M.S. Bayliss, 1999; Eisen et al., 1999; J. J. Gutteling et al., 2007; Yacavone et al., 2001; Younossi, 2001; Younossi & Guyatt, 1998). The advantage of using this instrument is that it can compare with other types of diseases or healthy control population (M.S. Bayliss, 1999; Eisen et al., 1999; J. J. Gutteling et al., 2007; Yacavone et al., 2001; Younossi, 2001; Younossi & Guyatt, 1998). Therefore, it is widely used in health services and comparative studies. However, generic measures may not detect small but important clinical changes specific to a particular patient group. The Medical Outcomes Study Short Form-36 (SF-36) is the most commonly used (J. J. Gutteling et al., 2007; Yacavone et al., 2001; Younossi, 2001; Younossi & Guyatt, 1998). It showed in a study by Foster et al. that CHB patients had significant lower HRQOL scores in mental health and general health perception aspects, but their physical related HRQOL scores were comparable to the healthy control (Foster et al., 1998). Previous studies on the use of SF-36 on CHB patients have demonstrated that patients with less severe disease had lower HRQOL scores in general health compared to those with general population or healthy controls (Bondini et al., 2007; Foster et al., 1998; S C Ong et al., 2008; Tan et al., 2008). Once they developed complications, lower HRQOL scores was found in both physical and mental health (S C Ong et al., 2008). The effectiveness of anti-viral treatment was detected by the SF-36 in a longitudinal study on 150 Chinese CHB patients at different stages of illness

A disease-specific measure theoretically can detect small but clinically important changes on HRQOL that are unique to the particular condition although it does not allow for comparison with the general population or other disease groups (M.S. Bayliss, 1999; Eisen et al., 1999; Younossi, 2001; J. J. Gutteling et al., 2007; Yacavone et al., 2001). Several HRQOL measures specific for chronic liver disease patients, such as the Chronic Liver Disease Questionnaire (CLDQ) (Younossi, Guyatt, Kiwi, Boparai, & King, 1999), the Hepatitis Quality of Life Questionnaire (HQLQ) (M. S. Bayliss et al., 1998), the Liver Disease Quality of Life Questionnaire (LDQOL) (Gralnek et al., 2000), the Liver Disease Symptom Index (LDSI 1.0 and 2.0) (Unal et al., 2001; van der Plas et al., 2004), the Hepatitis B Quality of Life (HBQOL) (Spiegel et al., 2007) and Chronic Liver Disease-Specific Quality of Life (CLD-QOL) (Lee et al., 2007) are available. Each instrument has its

lower mental HRQOL scores (Sobhonslidsuk et al., 2006).

**2.3.2 HRQOL measures applicable to CHB** 

receiving LVD treatment (Yi, 2006).


advantages and disadvantages. Table 1 presents the characteristics of these diseasespecific HRQOL measures.

Table 1. Characteristics of Disease-specific HRQOL Measures

The Chronic Liver Disease Questionnaire (CLDQ) is the first disease-specific HRQOL measure for evaluating patients with chronic liver disease (CLD) developed by Younossi et al (Younossi et al., 1999). The CLDQ has 29 items generated by patients with chronic liver disease, hepatologists, and a review of literature (Younossi et al., 1999). The CLDQ has six scales measuring fatigue, activity, emotional function, abdominal symptoms, systemic symptoms and worry (Younossi et al., 1999), which captures the important problems associated with CHB infection and its complications. It is scored with six domain and one summary scores (Younossi et al., 1999). This short measure can be completed in less than 15 minutes, a criterion for assuring a good response rate (Cella & Tulsky, 1990; McColl,

Reviews on the Epidemiology, Quality of Life, and Management of Chronic Hepatitis B (CHB) 391

questionnaire including items assessing psychological well-being, anticipation anxiety, vitality, disease stigma, vulnerability, and transmissibility (Spiegel et al., 2007). It has shown to be valid and reliable in English-speaking patients in the United States (Spiegel et al., 2007). The validity and applicability of this instrument on CHB patients with complications

The Chronic Liver Disease-Quality of Life questionnaire (CLD-QOL) was designed to measure HRQOL of Asian patients with chronic liver disease, the first of its kind (Lee et al., 2007). There are 27 items which are organized into the domains of specific symptoms, social function, emotional status, general symptoms and uncertain future (Lee et al., 2007). Lee et al found a significant difference in HRQOL scores between patients with mild stage of cirrhosis and moderate to severe stage of cirrhosis, supporting construct validity of CLD-QOL (Lee et al., 2007). Further evaluation on its psychometric properties on patients

Cost-effectiveness analysis (CEA) is an area of increasing interest among researchers, physicians and policy makers (Sun et al., 2007). CEA is a method of summarizing the health benefits and resources used by health programmes, therefore policy makers can select among them (Russell, Gold, Siegel, Daniels, & Weinstein, 1996; Weinstein, 1990; Weinstein, Siegel, Gold, Kamlet, & Russell, 1996). It summarizes all programme costs and benefits (effectiveness), and uses economic theory to aid choice between competing health programmes when resources are scarce (Russell et al., 1996; Weinstein, 1990; Weinstein et al., 1996). CEA can express health benefits in more generic terms, such as quality adjusted life years (QALYs) gained (Weinstein, 1990). It provides a common unit to allow comparisons between different disease groups or intervention programmes (Weinstein, Torrance, & McGuire, 2009). This method is particularly useful in the analysis of preventive health programmes, such as anti-viral treatment in CHB patients (Kanwal et al., 2005;

Most HRQOL measures give a profile of domain scores that are not designed for economic evaluation. Treatment evaluation by profile scores may give inconsistent results and lead to a piecemeal understanding of the impact of an intervention because of variations in the effect on different domains. For example, one scale indicates a beneficial effect whereas other scales may give negative results. Therefore, there is a need for methods to combine multidimensional information in more systematic ways. Research has shown that multidimensional HRQOL states can be converted to a composite preference value expressed on a numerical scale ranging from 0 (death) to 1 (perfect health) (Brazier et al., 2002), based on preference valuation by subjects from the general population (Guide to the methods of technology appraisal). It is possible to have negative preference for states that are worse

An important application of preference based measures is a measure of effectiveness in health economic evaluation (Preedy et al., 2010). Quality adjusted life year (QALY) is increasingly used as a measure of health outcomes in the past 20 years (Neumann, Greenberg, Olchanski, Stone, & Rosen, 2005). QALYs give a single index combining morbidity and mortality and is easy to calculate if the preference of health status is known

without cirrhosis is needed before the instrument can be applied more widely.

**2.3.3 Preference-based measure (utility) of HRQOL** 

Sullivan et al., 2007; Sun et al., 2007; Yuan et al., 2008).

than being dead (Preedy et al., 2010).

and other cultures are not known.

Christiansen, & Konig-Zahn, 1997). Patients with different types and stages of liver disease were included in the development and validation process supporting its broad application in hepatology research (Younossi et al., 1999). It has been shown to have adequate internal reliability, validity and sensitivity (Younossi et al., 1999). The CLDQ has been shown to be suitable for cross-cultural adaptation to different cultures (Ferrer et al., 2006; Hauser, Schnur, et al., 2004; Rucci et al., 2005; Sobhonslidsuk, Silpakit, Kongsakon, Satitpornkul, & Sripetch, 2004; Wu, Deng, Ji, & Yan, 2003) including Italian, German, Chinese (Mainland) and Thai. Recently, it has also been translated into Portuguese and Bengali (Mucci, Citero Vde, Gonzalez, De Marco, & Nogueira-Martins, 2010; Ray, Dutta, Basu, & De, 2010). However, the responsiveness of the CLDQ has not been investigated widely. Further research is needed to demonstrate its ability to detect change over time or with intervention.

The other liver disease specific HRQOL measures are less widely used because they are either much longer or have relatively few data supporting their validity or sensitivity. The HQLQ developed by Bayliss has 69 items combining the generic SF-36 scales with three additional generic scales (positive well-being, sleep and health distress) and two hepatitis C specific scales (health distress and limitations because of hepatitis C ) and was intended for patients with chronic hepatitis C (CHC) infection (M. S. Bayliss et al., 1998). It has been shown to be sensitive in patients with CHC but data on patients with CHB infection are few (M. S. Bayliss et al., 1998). The instrument was recently translated and validated in patients with CHB in Singapore (S. C. Ong, Lim, & Li, 2009, 2010). The main disadvantage of the HQLQ is many liver disease specific symptoms, such as abdominal pain, are not addressed despite its length (M. S. Bayliss et al., 1998). Furthermore, significant ceiling effects of three scales were observed (M. S. Bayliss et al., 1998), and the instrument's responsiveness remains unknown (S. C. Ong et al., 2009, 2010).

The Liver Disease Quality of Life instrument (LDQOL) was developed by Gralnek et al consisting of generic and disease-specific scales with a total of 101 items (Gralnek et al., 2000). It is not very widely used because its length limits its acceptability. It is applicable mainly to patients with advanced liver disease or waiting for liver transplantation (Casanovas et al., 2003; Dias Teixeira, de Fatima Gomes de Sa Ribeiro, & Strauss, 2005). In other words, it was not designed for patients with less severe liver disease. The LDQOL has been translated and adapted into Spanish and Catalan in transplant patients as well as Brazilian Portuguese in patients with chronic liver disease (Casanovas et al., 2003; Dias Teixeira et al., 2005). Pilot testing has supported the reliability and validity of the LDQOL but it has not been tested in longitudinal studies (Dias Teixeira et al., 2005; Gralnek et al., 2000). Recently, a short form (36 items) of the liver disease quality of life instrument (SF-LDQOL) was validated on patients with advanced liver disease (Kanwal et al., 2008). However, its validity may not be generalizable to patients with asymptomatic hepatitis B or an early stage of liver disease (Kanwal et al., 2008).

The Liver Disease Symptom Index (LDSI) is a short instrument that consists of 18 items that measure nine disease-specific symptoms and the hindrance that patients experience from these symptoms (Unal et al., 2001; van der Plas et al., 2004). It has been validated on 374 patients but the data on its validity and other psychometric properties on CHB patients are limited (Unal et al., 2001).

The Hepatitis B Quality of Life instrument (HBQOL) is the first HRQOL measures designed for specifically hepatitis B patients without cirrhosis (Spiegel et al., 2007). It is a 31 items questionnaire including items assessing psychological well-being, anticipation anxiety, vitality, disease stigma, vulnerability, and transmissibility (Spiegel et al., 2007). It has shown to be valid and reliable in English-speaking patients in the United States (Spiegel et al., 2007). The validity and applicability of this instrument on CHB patients with complications and other cultures are not known.

The Chronic Liver Disease-Quality of Life questionnaire (CLD-QOL) was designed to measure HRQOL of Asian patients with chronic liver disease, the first of its kind (Lee et al., 2007). There are 27 items which are organized into the domains of specific symptoms, social function, emotional status, general symptoms and uncertain future (Lee et al., 2007). Lee et al found a significant difference in HRQOL scores between patients with mild stage of cirrhosis and moderate to severe stage of cirrhosis, supporting construct validity of CLD-QOL (Lee et al., 2007). Further evaluation on its psychometric properties on patients without cirrhosis is needed before the instrument can be applied more widely.

#### **2.3.3 Preference-based measure (utility) of HRQOL**

390 Hepatocellular Carcinoma – Basic Research

Christiansen, & Konig-Zahn, 1997). Patients with different types and stages of liver disease were included in the development and validation process supporting its broad application in hepatology research (Younossi et al., 1999). It has been shown to have adequate internal reliability, validity and sensitivity (Younossi et al., 1999). The CLDQ has been shown to be suitable for cross-cultural adaptation to different cultures (Ferrer et al., 2006; Hauser, Schnur, et al., 2004; Rucci et al., 2005; Sobhonslidsuk, Silpakit, Kongsakon, Satitpornkul, & Sripetch, 2004; Wu, Deng, Ji, & Yan, 2003) including Italian, German, Chinese (Mainland) and Thai. Recently, it has also been translated into Portuguese and Bengali (Mucci, Citero Vde, Gonzalez, De Marco, & Nogueira-Martins, 2010; Ray, Dutta, Basu, & De, 2010). However, the responsiveness of the CLDQ has not been investigated widely. Further research is needed to demonstrate its ability to detect change over time or with intervention. The other liver disease specific HRQOL measures are less widely used because they are either much longer or have relatively few data supporting their validity or sensitivity. The HQLQ developed by Bayliss has 69 items combining the generic SF-36 scales with three additional generic scales (positive well-being, sleep and health distress) and two hepatitis C specific scales (health distress and limitations because of hepatitis C ) and was intended for patients with chronic hepatitis C (CHC) infection (M. S. Bayliss et al., 1998). It has been shown to be sensitive in patients with CHC but data on patients with CHB infection are few (M. S. Bayliss et al., 1998). The instrument was recently translated and validated in patients with CHB in Singapore (S. C. Ong, Lim, & Li, 2009, 2010). The main disadvantage of the HQLQ is many liver disease specific symptoms, such as abdominal pain, are not addressed despite its length (M. S. Bayliss et al., 1998). Furthermore, significant ceiling effects of three scales were observed (M. S. Bayliss et al., 1998), and the instrument's responsiveness

The Liver Disease Quality of Life instrument (LDQOL) was developed by Gralnek et al consisting of generic and disease-specific scales with a total of 101 items (Gralnek et al., 2000). It is not very widely used because its length limits its acceptability. It is applicable mainly to patients with advanced liver disease or waiting for liver transplantation (Casanovas et al., 2003; Dias Teixeira, de Fatima Gomes de Sa Ribeiro, & Strauss, 2005). In other words, it was not designed for patients with less severe liver disease. The LDQOL has been translated and adapted into Spanish and Catalan in transplant patients as well as Brazilian Portuguese in patients with chronic liver disease (Casanovas et al., 2003; Dias Teixeira et al., 2005). Pilot testing has supported the reliability and validity of the LDQOL but it has not been tested in longitudinal studies (Dias Teixeira et al., 2005; Gralnek et al., 2000). Recently, a short form (36 items) of the liver disease quality of life instrument (SF-LDQOL) was validated on patients with advanced liver disease (Kanwal et al., 2008). However, its validity may not be generalizable to patients with asymptomatic hepatitis B or

The Liver Disease Symptom Index (LDSI) is a short instrument that consists of 18 items that measure nine disease-specific symptoms and the hindrance that patients experience from these symptoms (Unal et al., 2001; van der Plas et al., 2004). It has been validated on 374 patients but the data on its validity and other psychometric properties on CHB patients are

The Hepatitis B Quality of Life instrument (HBQOL) is the first HRQOL measures designed for specifically hepatitis B patients without cirrhosis (Spiegel et al., 2007). It is a 31 items

remains unknown (S. C. Ong et al., 2009, 2010).

an early stage of liver disease (Kanwal et al., 2008).

limited (Unal et al., 2001).

Cost-effectiveness analysis (CEA) is an area of increasing interest among researchers, physicians and policy makers (Sun et al., 2007). CEA is a method of summarizing the health benefits and resources used by health programmes, therefore policy makers can select among them (Russell, Gold, Siegel, Daniels, & Weinstein, 1996; Weinstein, 1990; Weinstein, Siegel, Gold, Kamlet, & Russell, 1996). It summarizes all programme costs and benefits (effectiveness), and uses economic theory to aid choice between competing health programmes when resources are scarce (Russell et al., 1996; Weinstein, 1990; Weinstein et al., 1996). CEA can express health benefits in more generic terms, such as quality adjusted life years (QALYs) gained (Weinstein, 1990). It provides a common unit to allow comparisons between different disease groups or intervention programmes (Weinstein, Torrance, & McGuire, 2009). This method is particularly useful in the analysis of preventive health programmes, such as anti-viral treatment in CHB patients (Kanwal et al., 2005; Sullivan et al., 2007; Sun et al., 2007; Yuan et al., 2008).

Most HRQOL measures give a profile of domain scores that are not designed for economic evaluation. Treatment evaluation by profile scores may give inconsistent results and lead to a piecemeal understanding of the impact of an intervention because of variations in the effect on different domains. For example, one scale indicates a beneficial effect whereas other scales may give negative results. Therefore, there is a need for methods to combine multidimensional information in more systematic ways. Research has shown that multidimensional HRQOL states can be converted to a composite preference value expressed on a numerical scale ranging from 0 (death) to 1 (perfect health) (Brazier et al., 2002), based on preference valuation by subjects from the general population (Guide to the methods of technology appraisal). It is possible to have negative preference for states that are worse than being dead (Preedy et al., 2010).

An important application of preference based measures is a measure of effectiveness in health economic evaluation (Preedy et al., 2010). Quality adjusted life year (QALY) is increasingly used as a measure of health outcomes in the past 20 years (Neumann, Greenberg, Olchanski, Stone, & Rosen, 2005). QALYs give a single index combining morbidity and mortality and is easy to calculate if the preference of health status is known

Reviews on the Epidemiology, Quality of Life, and Management of Chronic Hepatitis B (CHB) 393

hepatocellular carcinoma (HCC) had very low preference values (Levy et al., 2008), indicating a strong impact of CHB. The results of Levy's study had limitations because it rated preference values on disease-specific health states (Levy et al., 2008), which is in contrary from NICE's recommendation that health preference should be measured by generic measures based on valuations by the general public (Guide to the methods of technology appraisal). The other study was conducted in Singapore by Ong et al, which found significantly lower health preference values measured by EQ-5D in patients with DC and HCC compared with asymptomatic hepatitis B (AHB) carriers (S C Ong et al., 2008). A small study on 140 patients with chronic liver diseases, with 36% having CHB found that patients with CHB had better health preference values than patients with other liver

More studies on health preference values associated with each stage of CHB from AHB, to CHB with ILF, cirrhosis and HCC, using locally validated preference-based measures, are needed to provide more accurate estimates of the change in preference value with disease progression. Such information is useful for the evaluation the cost-effectiveness of anti-

Hepatitis B is a significant health problem in Southern China and Hong Kong. Chronic hepatitis B (CHB) is a chronic disease that puts significant demand on health services. Regular monitoring is needed for progression of disease and development of complications. Anti-viral treatments may be needed for the eradication of the virus and other interventions are required when complications develop. Understanding the impact of illness on quality of life can make health care more responsive to patients' needs. Studies have shown significant lower health-related quality of life (HRQOL) scores in CHB patients especially in the presence of complications. Data on HRQOL in patients with CHB are limited. Generic and disease specific measures can complement each other in the evaluation of the impact of CHB on HRQOL. The Chronic Liver Disease Questionnaire (CLDQ) is a liver disease-specific HRQOL measure applicable to patients with chronic liver disease at different stages of illness. It has the best face validity for adaptation to be used in Chinese patients with CHB in Hong Kong. There are lots of potential applications of HRQOL data of CHB patients. It can inform policy and practice to make health service more patient-centered. HRQOL may also be used as an outcome measure of effectiveness of treatment and quality of care. HRQOL can be converted into a preference value for the calculation of quality adjusted life year

Afendy, A., et al. (2009). Predictors of health-related quality of life in patients with chronic

Ayoub, W. S., & Keeffe, E. B. (2008). Review article: current antiviral therapy of chronic

Bayliss, M. S. (1999). Methods in outcomes research in hepatology: definitions and domains

(QALYs) in cost-effectiveness analysis of medical intervention.

hepatitis B. *Aliment Pharmacol Ther*.

liver disease. *Aliment Pharmacol Ther, 30*(5), 469-476.

of quality of life. *Hepatology, 29 (Supp 1)*, 3S-6S.

diseases (A. A. Dan et al., 2008).

viral treatments.

**3. Conclusion** 

**4. References** 

(Weinstein et al., 2009). It provides a common metric to compare with different treatment options with the other, to compare treatment side-effects versus benefit, or to compare intervention programs with another one (Preedy et al., 2010; Weinstein et al., 2009). QALYs have been applied to different diseases or intervention programs, for instance, treatment of coronary heart disease and screening for breast cancer (Chan, Nallamothu, Gurm, Hayward, & Vijan, 2007; Wong, Kuntz, Cowling, Lam, & Leung, 2007). Information on the preference values of different CHB states can be combined with life years gain in the evaluation of the cost-effectiveness of different anti-viral treatment strategies for CHB (Kanwal et al., 2005; Sullivan et al., 2007; Veenstra et al., 2008; Yuan et al., 2008).

Three most commonly used preference based measures of health are the Short Form-6D (SF-6D), the Health Utilities Index (HUI) and the EuroQol EQ-5D (Brazier et al., 2002; Brooks, 1996; Feeny, Furlong, Boyle, & Torrance, 1995). They are most useful for the evaluation of the cost-effectiveness of anti-viral treatment for CHB. Table 2 summarizes the domains of the commonly used generic HRQOL measures including the SF-6D, HUI and EQ-5D. They may give different results and interpretations because of different dimensions and methods of preference valuation. Preference values are also population specific that scoring algorithms derived from America or Europe may not apply to the Chinese. The SF-6D is the only measure that has been translated for and validated in the Chinese population in Hong Kong (C. L. K. Lam, Brazier, & McGhee, 2008).


Table 2. The Comparison of Content Domains among Different HRQOL Measures

Several studies have assessed health preference in chronic liver diseases, but the majority of them were on chronic hepatitis C (CHC) (Chong et al., 2003; McLernon, Dillon, & Donnan, 2008; Thein, Krahn, Kaldor, & Dore, 2005). Only two large studies evaluating health preference of CHB values were found in the literature (Levy et al., 2008; S C Ong et al., 2008). The first one was a multi-center study to elicit utilities for six hypothetical states from infected and uninfected individuals by Levy et al (Levy et al., 2008). The results found that health preference values declined with disease progressing from 0.68 in uncomplicated CHB to 0.35 in decompensated cirrhosis (DC) (Levy et al., 2008). Patients with DC and hepatocellular carcinoma (HCC) had very low preference values (Levy et al., 2008), indicating a strong impact of CHB. The results of Levy's study had limitations because it rated preference values on disease-specific health states (Levy et al., 2008), which is in contrary from NICE's recommendation that health preference should be measured by generic measures based on valuations by the general public (Guide to the methods of technology appraisal). The other study was conducted in Singapore by Ong et al, which found significantly lower health preference values measured by EQ-5D in patients with DC and HCC compared with asymptomatic hepatitis B (AHB) carriers (S C Ong et al., 2008). A small study on 140 patients with chronic liver diseases, with 36% having CHB found that patients with CHB had better health preference values than patients with other liver diseases (A. A. Dan et al., 2008).

More studies on health preference values associated with each stage of CHB from AHB, to CHB with ILF, cirrhosis and HCC, using locally validated preference-based measures, are needed to provide more accurate estimates of the change in preference value with disease progression. Such information is useful for the evaluation the cost-effectiveness of antiviral treatments.

#### **3. Conclusion**

392 Hepatocellular Carcinoma – Basic Research

(Weinstein et al., 2009). It provides a common metric to compare with different treatment options with the other, to compare treatment side-effects versus benefit, or to compare intervention programs with another one (Preedy et al., 2010; Weinstein et al., 2009). QALYs have been applied to different diseases or intervention programs, for instance, treatment of coronary heart disease and screening for breast cancer (Chan, Nallamothu, Gurm, Hayward, & Vijan, 2007; Wong, Kuntz, Cowling, Lam, & Leung, 2007). Information on the preference values of different CHB states can be combined with life years gain in the evaluation of the cost-effectiveness of different anti-viral treatment strategies for CHB (Kanwal et al., 2005;

Three most commonly used preference based measures of health are the Short Form-6D (SF-6D), the Health Utilities Index (HUI) and the EuroQol EQ-5D (Brazier et al., 2002; Brooks, 1996; Feeny, Furlong, Boyle, & Torrance, 1995). They are most useful for the evaluation of the cost-effectiveness of anti-viral treatment for CHB. Table 2 summarizes the domains of the commonly used generic HRQOL measures including the SF-6D, HUI and EQ-5D. They may give different results and interpretations because of different dimensions and methods of preference valuation. Preference values are also population specific that scoring algorithms derived from America or Europe may not apply to the Chinese. The SF-6D is the only measure that has been translated for and validated in the Chinese population in Hong

 SF-6D HUI EQ-5D HRQOL Utility Utility Utility

Year 2002 1995 1990
