复旦大学：《循证医学》课程教学资源（预习资料）生命质量分析评价预习.pdf

Unit 8：生命质量的分析评价主讲教师：姜林娣助理教师：马莉莉授课时间：2010 年 4 月 13 日（1 班）；2010 年 4 月 16 日（2 班）一、教学目的：掌握和熟悉生命质量评价的概念和方法二、教学内容：1、掌握：生命质量分析评价的方法和原理；2、熟悉：生命质量分析的应用,效用值临床意义，测定方法；3、了解：生命质量分析的常用量表三、教学重点：生命质量分析评价的方法和原理四、教学难点：1、生命质量分析评价方法的选择及对临床决策的作用。2、效用值测定方法五、课堂知识点： 1．功能状况、健康和生命质量 2．健康相关生活质量包含内容 3．健康相关生活质量量表主要检验方面 4．效用值定义、表示方法和测定方法六、中文和英文关键词生命质量分析 quality of life analysis，质量调整生命年，quality-adjusted life year (QALY)，费用-效果评价 cost-effectiveness analysis 七、预习与思考： 1、The burden of osteoarthritis: clinical and quality-of-life issues.Moskowitz RW (1)关节有哪些损害 (2) 我们如何评价它 (3)临床上还有那些实例：如肿瘤、心血管疾病 2、SF-36 健康调查量表中文版的研制及其性能测试（李鲁） (1)熟悉量表检验包括几部分，几个方面如何将国外量表应用到我们的临床工作中？ (2)国外量表应用于临床的过程？ 3、A Systematic Review of Health State Utility Values for Osteoporosis Related Conditions (J. E. Brazier1) (1)效用值临床应用价值 (2)测定效用值方法 (3)测定效用值的人群 4、EQ-5D 量表和 HAQ 量表八、参考书及文献目录 1.《循证医学与临床实践》（第 2 版），王吉耀主编，科学出版社 2. The Lancet handbook of essential concepts in clinical research. Schulz KF, Grimes DA. Philadelphia, PA, USA: Elsevier, 2006. 3. Measuring quality of life using quality of life measures in the clinical setting. I.Higginson et al. BMJ 2001;322:1297-1300 . 4. www.sf-36.org 第 131 页

REPORT E The Burden of osteoarthritis: Clinical and Quality-of-Life Issues Roland W. Moskowitz. MD O steoarthritis (OA)is the most common form of arthritis, affecting 27 million adults in the United Abstract States. OA typically occurs in the hands, knees, steoarthritis (OA), the most common form rados y of jolesepine, and hips, although it may be seen in any of a f arthritis, is a potentially devastating joint mally supported through the use of laboratory studies. Although OA degenerative process accompanied by low- is often characterized as a degenerative disease grade inflammation, and, although there is gra a strong correlation between age and OA mation actually constitutes an important aspect of OA's pathologic risk, the abnormal changes that occur in the articular cartilage of people with OA differ OA is strongly correlated with aging: the risk of OAVincreases con notably from the typical changes associated siderably with each decade after the age of about 45 years. Nevertheless with joint aging in several important ways aging is not inevitably associated with OA. In fact, several pathophysi Risk factors for OA are multiple and span a variety of risk domains, such as lifestyle issues(eg, obesity and engagement in that associated with age-related changes in cartilage. That said,such manual labor), genetic predisposition, sex age-related changes do play an important role in OA pathogenesis and and ethnicity (risk is higher in women and predispose individuals to the d African Americans), and comorbidities. -Other th ber of risk factors for Clinical outcomes for people with Oa typi OA, including comorbidities both related and unrelated to musculo. cally involve pain, limitations of daily living skeletal conditions. ' The presence of other joint diseases is the most activities, and overall diminution of qual common Inuse k factor, whereas obesity ity of life (QoL). The need to evaluate the degree of this burden, as well as to deter on nonmusculoskeletal comorbidities associat mine treatment approaches and measure ed with OA. 7 Lifestyle variables, such as a history of manual labor and their success, requires instruments for mea- cigarette smoking, as well as sex-and phenotype-related conditions suring QOL. The 2 most commonly used such at menarche and joint hyperlaxity instruments to measure ooL in OA are the a role in conferring risk of oA. The genetic component of OA risk Western Ontario and McMaster Universities Osteoarthritis Index(WOMAC) and the while still being studied at present, has been partially elucidated in Medical Outcomes Study 36-Item Short de scan stud ed genetic vari Form Health Survey. Both provide useful nts' associated with OA. ommunications, global information to the clinician and Ultimately, it is the burden of suffering experienced by people researcher alike about pain and function in with Oa that is of primary concern, and that burden can be signifi patients with OA, although the WOMAC is more often used in the clinical setting as cant Pain and functional impairment are the key domains of that it is self-administered. A number of other burden,and taken together they often exert a significant reduction pain and function-specific measures are in quality of life(QOL). 0-13 The present review will briefly describe also available that may provide additiona the pathophysiology, prevalence, and typical outcomes of Oa before insight into patient status when used in addressing the issue of QoL in Oa and the best means in which to combination with global QoL instruments measure It Am J Manag Care. 2009: 15: S223-S229) For author information and disclosures. see end of text. Pathophysiology of OA Cartilage remodeling involves balanced interactions of synthesis VOL 15. NO. 8 THE AMERICAL第u13L

© Managed Care & Healthcare Communications, LLC VOL. 15, No. 8 n The American Journal of Managed Care n S223 Osteoarthritis (OA) is the most common form of arthritis, affecting 27 million adults in the United States.1 OA typically occurs in the hands, knees, spine, and hips, although it may be seen in any of a variety of joints.2 Clinical diagnosis is based on observed symptoms, radiographic changes, or both, whereas differential diagnosis is normally supported through the use of laboratory studies. Although OA is often characterized as a degenerative disease, low-grade inflammation actually constitutes an important aspect of OA’s pathologic process.3,4 OA is strongly correlated with aging: the risk of OA increases considerably with each decade after the age of about 45 years.1 Nevertheless, aging is not inevitably associated with OA. In fact, several pathophysiologic changes that occur in osteoarthritic cartilage differ notably from that associated with age-related changes in cartilage.5 That said, such age-related changes do play an important role in OA pathogenesis and, at a minimum, predispose individuals to the disease.6 Other than increasing age, there are a number of risk factors for OA, including comorbidities both related and unrelated to musculoskeletal conditions.7 The presence of other joint diseases is the most common musculoskeletal comorbid risk factor, whereas obesity is among the most common nonmusculoskeletal comorbidities associated with OA.7 Lifestyle variables, such as a history of manual labor and cigarette smoking, as well as sex- and phenotype-related conditions— such as age at menarche and joint hyperlaxity in men—can also play a role in conferring risk of OA.8 The genetic component of OA risk, while still being studied at present, has been partially elucidated in recent years as genome-wide scan studies have identified genetic variants associated with OA.9 Ultimately, it is the burden of suffering experienced by people with OA that is of primary concern, and that burden can be significant. Pain and functional impairment are the key domains of that burden, and taken together they often exert a significant reduction in quality of life (QOL).10-13 The present review will briefly describe the pathophysiology, prevalence, and typical outcomes of OA before addressing the issue of QOL in OA and the best means in which to measure it. Pathophysiology of OA Cartilage remodeling involves balanced interactions of synthesis The Burden of Osteoarthritis: Clinical and Quality-of-Life Issues Roland W. Moskowitz, MD n report n Abstract Osteoarthritis (OA), the most common form of arthritis, is a potentially devastating joint disease, affecting 27 million US adults. Its pathophysiology is marked by a gradual degenerative process accompanied by lowgrade inflammation, and, although there is a strong correlation between age and OA risk, the abnormal changes that occur in the articular cartilage of people with OA differ notably from the typical changes associated with joint aging in several important ways. Risk factors for OA are multiple and span a variety of risk domains, such as lifestyle issues (eg, obesity and engagement in manual labor), genetic predisposition, sex and ethnicity (risk is higher in women and African Americans), and comorbidities. Clinical outcomes for people with OA typically involve pain, limitations of daily living activities, and overall diminution of quality of life (QOL). The need to evaluate the degree of this burden, as well as to determine treatment approaches and measure their success, requires instruments for measuring QOL. The 2 most commonly used instruments to measure QOL in OA are the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and the Medical Outcomes Study 36-Item ShortForm Health Survey. Both provide useful global information to the clinician and researcher alike about pain and function in patients with OA, although the WOMAC is more often used in the clinical setting as it is self-administered. A number of other pain and function-specific measures are also available that may provide additional insight into patient status when used in combination with global QOL instruments. (Am J Manag Care. 2009;15:S223-S229) For author information and disclosures, see end of text. 第 132 页

and degradation to achieve homeostasis of the ly doubled in patients aged 45 years or older extracellular matrix (ECM). 4 In Oa this pro- compared with those 26 years or older. The pro. cess becomes unbalanced, leading to pathologic portion of women with Oa also increases relative hanges in the affected joint. The articular carti- to men as age increases. Whereas the framingham lage cells, chrondocytes, are responsible for main- study found that 4.9% of women at least 26 years of taining homeostasis of the eCm by producing its age had knee Oa compared with 4.6% of men, the major components, collagen and proteoglycan, in gap increased to 7. 2% versus 5.9%, respectively, in response to deterioration. Changes in the chron- the 45 years or older group. This gap was replicated docytes are associated with abnormal anabolic and in the Johnston County study, although a highe catabolic activities as well as abnormal prolifera- rate of knee Oa for both men and women aged tion and apoptosis 45 years or older was observed. In that study, the In the early stages of OA, loosening of the col- rate of knee OA was 18.7% for women compared lagen network as well as proteoglycan loss occur with 13.5% for men. The Johnston County study in the upper cartilage zones and may still, at that further observed a higher rate of hip OA in women point, be reversible. 5 Over time, these changes 45 years or older(9.3%)compared with men in the occur within deeper cartilage zones, reducing the same age group(8.7%) lasticity of the cartilage and making a return The observation that the OA gap increases to homeostasis increasingly difficult to achieve. between men and women as they age is consis- Chrondocyte senescence--which is associated tent with incidence data from the framingham ith increasing age--also appears to play a part in Osteoarthritis Study focusing on individuals aged 63 a reduced capacity for cartilage repair and contrib- to 91 years(mean age, 70.8 years). 8 Among these utes to Oa progressed older subjects, the age-adjusted relative risk(rr)of Recent data support the notion that changes women experiencing radiographically determined in subchondral bone are also a factor in cartilage knee Oa compared with men was 1.79(95%con- degradation. The subchondral bone, which is in fidence interval [CIl, 1.08-2.94) s Symptomatic immediate proximity of cartilage, may contribute knee OA was almost twice as likely (rr, 1.96; to cytokines, growth factors, and prostaglandins 95% Cl, 1.01-3.82)for women compared with escalating-perhaps initiating-the degenerative men. 8 Estimates of the rate of increase for knee OA in women was approximately 2% per year fo radiographically determined disease s Prevalence Prevalence and Incidence of oa data regarding hand OA in more than 1000 study Collecting prevalence data from multiple subjects aged 71 to 100 years from the original sources, including the Third National Health and Framingham study (years analyzed: 1992-1993) Nutrition Examination Survey (NHANES Ill) n found a much higher rate of disease in women the Framingham Osteoarthritis Study, and the (26. 2%)compared with men(13.4%). The Figure Johnston County Osteoarthritis Project, the shows the distribution of Oa symptoms at various National Arthritis Data Workgroup arrived at joints in the hands for both men and women. 9 a prevalence figure for 2005 of 26.9 million Us Ethnicity also plays a role in OA risk. According adults(aged >25 years)with some form of OA. to prevalence data from NHANES III(1991-1994) This constitutes a growth of approximately 6 radiographic knee Oa was observed in 17.7% of nillion cases from 1995, more than a one-fourth African American participants aged 60 years or increase in just 10 years. This likely reflects, in part, older compared with 14.8% of Mexican American an aging of the US population, although increases and 11.9% of white participants in the same age in other related factors, such as obesity, as well group(both differences P <.01).20 Of note, the dual increases in methods of OA detection, may play a elevated risk of being both African American and role in this observed prevalence female was observed with a prevalence of radic Nevertheless, the prevalence ofoA does increase graphic knee OA of 60.2%(95%CL, 52.8-67.5).20 dramatically with age. Data from the Framingham Finally, with regard to the distribution of OA study showed that the prevalence of knee Oa across its 3 most common sites--hands, knees, and S224 w第4Rc SEPTEMBER 2009

Reports S224 n www.ajmc.com n september 2009 and degradation to achieve homeostasis of the extracellular matrix (ECM).14 In OA this process becomes unbalanced, leading to pathologic changes in the affected joint.15 The articular cartilage cells, chrondocytes, are responsible for maintaining homeostasis of the ECM by producing its major components, collagen and proteoglycan, in response to deterioration. Changes in the chrondocytes are associated with abnormal anabolic and catabolic activities as well as abnormal proliferation and apoptosis. In the early stages of OA, loosening of the collagen network as well as proteoglycan loss occur in the upper cartilage zones and may still, at that point, be reversible.15 Over time, these changes occur within deeper cartilage zones, reducing the elasticity of the cartilage and making a return to homeostasis increasingly difficult to achieve. Chrondocyte senescence—which is associated with increasing age—also appears to play a part in a reduced capacity for cartilage repair and contributes to OA progression.16 Recent data support the notion that changes in subchondral bone are also a factor in cartilage degradation.17 The subchondral bone, which is in immediate proximity of cartilage, may contribute to cytokines, growth factors, and prostaglandins escalating—perhaps initiating—the degenerative process. Prevalence and Incidence of OA Collecting prevalence data from multiple sources, including the Third National Health and Nutrition Examination Survey (NHANES III), the Framingham Osteoarthritis Study, and the Johnston County Osteoarthritis Project, the National Arthritis Data Workgroup arrived at a prevalence figure for 2005 of 26.9 million US adults (aged >25 years) with some form of OA.1 This constitutes a growth of approximately 6 million cases from 1995, more than a one-fourth increase in just 10 years. This likely reflects, in part, an aging of the US population, although increases in other related factors, such as obesity, as well as increases in methods of OA detection, may play a role in this observed prevalence. Nevertheless, the prevalence of OA does increase dramatically with age. Data from the Framingham study showed that the prevalence of knee OA nearly doubled in patients aged 45 years or older compared with those 26 years or older.1 The proportion of women with OA also increases relative to men as age increases. Whereas the Framingham study found that 4.9% of women at least 26 years of age had knee OA compared with 4.6% of men, the gap increased to 7.2% versus 5.9%, respectively, in the 45 years or older group. This gap was replicated in the Johnston County study, although a higher rate of knee OA for both men and women aged 45 years or older was observed.1 In that study, the rate of knee OA was 18.7% for women compared with 13.5% for men. The Johnston County study further observed a higher rate of hip OA in women 45 years or older (9.3%) compared with men in the same age group (8.7%). The observation that the OA gap increases between men and women as they age is consistent with incidence data from the Framingham Osteoarthritis Study focusing on individuals aged 63 to 91 years (mean age, 70.8 years).18 Among these older subjects, the age-adjusted relative risk (RR) of women experiencing radiographically determined knee OA compared with men was 1.79 (95% confidence interval [CI], 1.08-2.94).18 Symptomatic knee OA was almost twice as likely (RR, 1.96; 95% CI, 1.01-3.82) for women compared with men.18 Estimates of the rate of increase for knee OA in women was approximately 2% per year for radiographically determined disease.18 Prevalence data regarding hand OA in more than 1000 study subjects aged 71 to 100 years from the original Framingham study (years analyzed: 1992-1993) again found a much higher rate of disease in women (26.2%) compared with men (13.4%). The Figure shows the distribution of OA symptoms at various joints in the hands for both men and women.19 Ethnicity also plays a role in OA risk. According to prevalence data from NHANES III (1991-1994), radiographic knee OA was observed in 17.7% of African American participants aged 60 years or older compared with 14.8% of Mexican American and 11.9% of white participants in the same age group (both differences P <.01).20 Of note, the dual elevated risk of being both African American and female was observed with a prevalence of radiographic knee OA of 60.2% (95% CI, 52.8-67.5).20 Finally, with regard to the distribution of OA across its 3 most common sites—hands, knees, and 第 133 页

ahead of dementia, diabetes, prostate cancer, and need to understand its impact on Oa patients in cancer ta are consistent with order to guide decision making-to determine how those from the framingham study, which found that and what interventions are appropriate--in the among older (mean age -74 years)study partici- management of the disease. Measuring QOL pants, knee OA, taken alone, represented I of the 4 ther allows clinicians the opportunity to determine rgest causes of disability along with heart disease, the efficacy of a given intervention. Numerous depression, and stroke. A significant proportion instruments are currently available for measuring of the patient population in the framingham study different aspects of QOL in the Oa patient, includ- with knee Oa were unable to perform a variety ing those that measure general QOL, functional of activities of daily living, such as heavy home capacities, the experience of pain, and psychologi- chores (34% disabled), walking 1 mile(31%), stair cal dimensions of QOL. The most commonly used ing(10%), and grocery shopping(10%). In instrument specific to OA is the Western Ontario fact, the ability to walk I mile or to undertake light and McMaster Universities Osteoarthritis Index housekeeping was notably more restricted among (WOMAC). WOMAC is a 24-item self-report those with knee OA compared with matched questionnaire that addresses joint pain, stiffness, patients with heart disease. 3 and loss of function related to oa of the knee and It is interesting to note that the burden of Oa in hip 24-26 Since its initial validation, WOMAC has terms of functional deficits can affect areas beyond been widely used in clinical trials, and has been the immediate OA loci. For example, an analysis of repeatedly shown to provide utility as a measure participants in the Johnston County Osteoarthritis of patient QOL, response to treatment, prediction Project with joint-specific hand symptoms found that of treatment outcomes, as well as sensitivity to not only did they experience significant deterioration minimal perceptible clinical improvement. 4. An in performance-based functional status overall--as electronic touch-screen version of the WOMAC might be expected-but that the disability extended (the e-WOMAC) has shown similar responsiveness beyond that which one would intuitively associate to the paper version in Oa patients. 1 with hand OA. People with hand Oa symptoms in The Medical Outcomes Study 36-Item Short this study experienced significant deterioration in Form Health Survey (SF-36)provides an 8-scale performing both upper- and lower-extremity tasks. evaluation of physical and mental QOL based on 36 These results were based on both a self-report instru- questions. Although the SF-36 is not specific to ment(the Health Assessment Questionnaire [HAQI OA, it is, like WOMAC, widely used to guide OA Disability Index)as well as performance-based func- treatment in clinical trials, being fairly sensitive to tional measures(timed 5 chair stands"and gait minimal perceptible clinical improvement. 24.27429 32A neasured over an 8-foot walking course ) I more OA-specific instrument, the SF-36 Arthritis- A separate analysis of participants in the Johnston Specific Health Index, was developed to better knee OA, and the severity of knee pain in particu- not been widely adopted. ation, but to date has County Osteoarthritis Project found that having target the Oa patient popul ar, was associated with a high degree of functional Although both the WoMAC and SF-36 instru- impairment. Even mild knee pain was strongly ments function relatively well in assessing various associated with disability in performing 16 of the QOL domains in OA, WOMAC may be more 20 upper- and lower-extremity tasks included in the responsive than the SF-36 instrument to detecting HaQ disability index. For those with moderate-to- changes in function. 4 In fact, the validity of both severe pain, significant disability was observed with instruments has been challenged. 5 Although it all 20 tasks in the HAQ index(all P<.001). 12 should be noted that WOMAC is widely accepted The common presence of comorbidities within for its ability to measure pain and functional deficits the Oa patient population exerts additional deleteri- in OA, Stratford et al have questioned its validity ous effects on both physical functioning and pain in measuring pain. According to their analysis, the factors that comprise pain evaluation are not Measuring QoL in OA alid, while they state that the pain scale as a whole The necessity of measuring QOL arises from a is not internally consistent. The authors suggest S226 w第4瓶5 SEPTEMBER 2009

Reports S226 n www.ajmc.com n september 2009 ahead of dementia, diabetes, prostate cancer, and breast cancer.22 These data are consistent with those from the Framingham study, which found that among older (mean age ~74 years) study participants, knee OA, taken alone, represented 1 of the 4 largest causes of disability along with heart disease, depression, and stroke.13 A significant proportion of the patient population in the Framingham study with knee OA were unable to perform a variety of activities of daily living, such as heavy home chores (34% disabled), walking 1 mile (31%), stair climbing (10%), and grocery shopping (10%).13 In fact, the ability to walk 1 mile or to undertake light housekeeping was notably more restricted among those with knee OA compared with matched patients with heart disease.13 It is interesting to note that the burden of OA in terms of functional deficits can affect areas beyond the immediate OA loci. For example, an analysis of participants in the Johnston County Osteoarthritis Project with joint-specific hand symptoms found that not only did they experience significant deterioration in performance-based functional status overall—as might be expected—but that the disability extended beyond that which one would intuitively associate with hand OA. People with hand OA symptoms in this study experienced significant deterioration in performing both upper- and lower-extremity tasks.11 These results were based on both a self-report instrument (the Health Assessment Questionnaire [HAQ] Disability Index) as well as performance-based functional measures (timed “5 chair stands” and gait measured over an 8-foot walking course).11 A separate analysis of participants in the Johnston County Osteoarthritis Project found that having knee OA, and the severity of knee pain in particular, was associated with a high degree of functional impairment.12 Even mild knee pain was strongly associated with disability in performing 16 of the 20 upper- and lower-extremity tasks included in the HAQ disability index.12 For those with moderate-tosevere pain, significant disability was observed with all 20 tasks in the HAQ index (all P <.001).12 The common presence of comorbidities within the OA patient population exerts additional deleterious effects on both physical functioning and pain.10,23 Measuring QOL in OA The necessity of measuring QOL arises from a need to understand its impact on OA patients in order to guide decision making—to determine how and what interventions are appropriate—in the management of the disease. Measuring QOL further allows clinicians the opportunity to determine the efficacy of a given intervention. Numerous instruments are currently available for measuring different aspects of QOL in the OA patient, including those that measure general QOL, functional capacities, the experience of pain, and psychological dimensions of QOL. The most commonly used instrument specific to OA is the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC). WOMAC is a 24-item self-report questionnaire that addresses joint pain, stiffness, and loss of function related to OA of the knee and hip.24-26 Since its initial validation, WOMAC has been widely used in clinical trials, and has been repeatedly shown to provide utility as a measure of patient QOL, response to treatment, prediction of treatment outcomes, as well as sensitivity to minimal perceptible clinical improvement.24,27-30 An electronic touch-screen version of the WOMAC (the e-WOMAC) has shown similar responsiveness to the paper version in OA patients.31 The Medical Outcomes Study 36-Item Short Form Health Survey (SF-36) provides an 8-scale evaluation of physical and mental QOL based on 36 questions. Although the SF-36 is not specific to OA, it is, like WOMAC, widely used to guide OA treatment in clinical trials, being fairly sensitive to minimal perceptible clinical improvement.24,27-29,32 A more OA-specific instrument, the SF-36 ArthritisSpecific Health Index, was developed to better target the OA patient population, but to date has not been widely adopted.33,34 Although both the WOMAC and SF-36 instruments function relatively well in assessing various QOL domains in OA, WOMAC may be more responsive than the SF-36 instrument to detecting changes in function.24 In fact, the validity of both instruments has been challenged.35 Although it should be noted that WOMAC is widely accepted for its ability to measure pain and functional deficits in OA, Stratford et al have questioned its validity in measuring pain.35 According to their analysis, the factors that comprise pain evaluation are not valid, while they state that the pain scale as a whole is not internally consistent.35 The authors suggest 第 135 页