 |
 |
|
17 June 2003 | Volume 138 Issue 12 | Pages 960-968
Background: Guidelines are increasingly recommending preventive services starting at 50 years of age, and policymakers are considering such a recommendation for pneumococcal polysaccharide vaccination. The finding that pneumococcal vaccination is cost-saving for people 65 years of age or older raises the question of the vaccination's implications for other older adults, especially black people, whose disease incidence exceeds that of nonblack people, and those with high-risk conditions.
Objective: To assess the implications of vaccinating black and nonblack people 50 through 64 years of age against invasive pneumococcal disease.
Design: Cost-effectiveness analysis.
Data Sources: Published literature for vaccination effectiveness and cost estimates; data on disease incidence and case-fatality rates from the Centers for Disease Control and Prevention.
Target Population: Hypothetical cohort 50 through 64 years of age with the 1995 U.S. age distribution.
Time Horizon: Lifetime.
Perspective: Societal.
Intervention: Pneumococcal polysaccharide vaccination compared with no vaccination.
Outcome Measures: Incremental medical costs and health effects, in quality-adjusted life-years per vaccinee.
Results of Base-Case Analysis: Vaccination saved medical costs and improved health among high-risk black people ($27.55 savings per vaccinee) and nonblack people ($5.92 savings per vaccinee), excluding survivors' future costs. For low-risk black and nonblack people and the overall general population, vaccination cost $2477, $8195, and $3434, respectively, to gain 1 year of healthy life.
Results of Sensitivity Analysis: Excluding survivors' future costs, in the general immunocompetent population, cost per quality-adjusted life-year in global worst-case results ranged from $21 513 for black people to $68 871 for nonblack people; in the high-risk population, cost ranged from $11 548 for black people to $39 000 for nonblack people. In the global best case, vaccination was cost-saving for black and nonblack people in the general immunocompetent and high-risk populations, excluding survivors' future costs. The cost-effectiveness range was narrower in probabilistic sensitivity analyses, with 95% probabilistic intervals ranging from cost-saving to $1594 for black people and from cost-saving to $12 273 for nonblack people in the general immunocompetent population. Costs per quality-adjusted life-year for low-risk people with case-fatality rates from 1998 were $2477 for black people and $8195 for nonblack people, excluding survivors' medical costs.
Conclusions: These results support the current recommendation to vaccinate high-risk people and provide useful information for considering extending the recommendation to the general population 50 through 64 years of age. Lack of evidence about the effectiveness of revaccination for people 65 years of age or older, when disease risks are higher, argues for further research to guide vaccination policy.
Contribution
Implications
–The Editors
Policymakers are increasingly recommending age 50 years as a milestone for interventions to prevent disease (primary prevention) or to screen for asymptomatic disease for which treatment can prevent sickness and death (secondary prevention). Clinical guidelines have advised screening at age 50 years to detect early disease, such as breast and colorectal cancer, but only recently have begun to include vaccinations (1, 2). In 1999, the Advisory Committee on Immunization Practices, which advises the Centers for Disease Control and Prevention (CDC), recommended reducing the age for universal influenza vaccination from 65 years to 50 years.
The Advisory Committee on Immunization Practices has also considered lowering the recommended age for universal pneumococcal polysaccharide vaccination from 65 years to 50 years. A cost-effectiveness analysis concluded that pneumococcal vaccination to prevent bacteremia in people 65 years of age or older conveys health benefits and saves medical costs (3). This result raised the question of the advisability of vaccinating adults younger than 65 years in age, especially black people, whose incidence of invasive pneumococcal disease is two to five times higher than that in white people (4, 5). After childhood, the risk for illness and death from the disease increases with age (5). Incidence per 100 000 population in 1998 increased from 3.9 for people 5 through 17 years of age to 7.4 for people 18 through 34 years of age, 16.0 for people 35 through 49 years of age, 23.0 for people 50 through 64 years of age, and 59.7 for people 65 years of age or older. The case-fatality rate also increases with age, from 1.4% for people younger than 2 years of age to 20.6% for those 80 years of age or older. Multiple administrations of this polysaccharide vaccine may not produce good immune responses each time. Although data on the clinical effectiveness of revaccination are lacking, serologic studies suggest that immune responses are lower after a second dose of pneumococcal polysaccharide vaccine than after the first dose (6, 7).
The interest of the Advisory Committee on Immunization Practices in possibly lowering the recommended age for universal vaccination also stemmed from the low utilization rates in people with underlying conditions that put them at high risk for disease. Despite the recommendation that high-risk people younger than 65 years of age receive vaccination, cumulative vaccination rates have been low: Only about 20% of high-risk people 50 through 64 years of age and 12% of high-risk black people in this age group were vaccinated by 1995 (8). Perhaps vaccination would more effectively reach these high-risk people if the recommendation were based on age instead of underlying condition, since utilization rates for people 65 years of age or older, spurred by programs to promote universally recommended vaccination, increased to almost 55% by 1999 (9).
Given the epidemiology of invasive pneumococcal disease, limitations of the current adult vaccine, and inadequate vaccination for high-risk people, policy issues center on the 50- through 64-year-old age group. Using cost-effectiveness analysis, we compared the likely health benefits and medical costs of vaccination and no vaccination for the total general immunocompetent, high-risk, and low-risk populations, 50 through 64 years of age, with separate analyses for black and nonblack people. The analysis excluded potential benefits for immunocompromised people.
In a Markov decision-tree model, we followed two hypothetical U.S. cohorts of vaccinated and unvaccinated people 50 through 64 years of age until death for the probability that they would incur invasive pneumococcal disease, die, or have average health (Figure). We used DATA, version 3.5 (TreeAge Software, Inc., Williamstown, Massachusetts), for analysis. The incremental medical costs of vaccination in 1995 U.S. dollars compared with the unvaccinated cohort was the sum of the vaccine costs, its administration, and treatment of adverse effects, minus savings in treatment costs from preventing invasive pneumococcal disease. The incremental health effect in quality-adjusted life-years (QALYs) was the total of increased years of life from preventing invasive pneumococcal disease, minus reductions from less-than-perfect health during surviving years. The analysis took a societal perspective, which included costs regardless of payer; calculated health and cost implications through life expectancies; and discounted future costs and health effects to reflect preferences for present over future events (10, 11). We applied the model to the total general immunocompetent, high-risk, and low-risk populations. ARTICLE
Cost-Effectiveness of Vaccination against Invasive Pneumococcal Disease among People 50 through 64 Years of Age: Role of Comorbid Conditions and Race
Editors' Notes
Top
Editors' Notes
Methods
Results
Discussion
Author & Article Info
References
Context
Methods
Top
Editors' Notes
Methods
Results
Discussion
Author & Article Info
References
Cost-Effectiveness Model
|
Sensitivity Analyses
The base-case analysis, following current guidelines, includes only future medical costs related to invasive pneumococcal disease (10). Because related and unrelated costs may be difficult to distinguish, supplementary analyses added survivors' total future medical costs. In addition to the base-case analyses, which used the most probable values of each variable, one-way sensitivity analyses varied uncertain variables one at a time over reasonable ranges. In addition, global best-case and worst-case (multiway) analyses used the extreme values of these variables simultaneously. The best-case analysis used values of uncertain variables with the most favorable health and cost implications for vaccination, such as high incidence of invasive pneumococcal disease, case-fatality rates, vaccination effectiveness rates, and QALY weights, and the worst-case analysis used values least favorable for vaccination (Table 1). We also conducted probabilistic sensitivity analyses for black and nonblack people in the general immunocompetent population. Each variable in the model was represented by a triangular distribution, where the peak was equal to the base-case value and the lower and upper bounds were equal to the best- or worst-case values. We performed 1000 simulations to generate a distribution of cost-effectiveness ratios according to the cost-effectiveness plane (3).
|
Health Effects
For invasive pneumococcal disease incidence, cases were defined by isolation of Streptococcus pneumoniae from a normally sterile body site. Most isolates are from blood (>90%) or cerebrospinal fluid (5% to 10%). Pneumonia is reported for at least 58% of patients with invasive pneumococcal disease 18 through 64 years of age (19). If vaccination protects against nonbacteremic pneumonia as well as invasive pneumococcal disease, our results underestimate benefits gained.
General Immunocompetent Population
For incidence of invasive pneumococcal disease and case-fatality rates, the base-case analysis used average rates of laboratory-confirmed disease from CDC active surveillance among defined populations in three cities—Baltimore, Maryland; San Antonio, Texas; and San Francisco, California—during 1995 and 1996, by black or nonblack race and age group (Table 1). Data through 2000, which became available more recently, show no consistent pattern of change.
From a case–control study with the best available age-specific data for immunocompetent patients, we derived vaccination effectiveness in preventing invasive pneumococcal disease caused by vaccine serotypes and duration of immunity (Appendix) (12). On the basis of these results, we estimated a maximum protection duration of 6 years, after which vaccinated and unvaccinated people had the same risk for death (Table 1). Following current guidelines, the analysis extends through life expectancies (10).
We assumed that vaccination effectiveness and case-fatality are the same for all racial or ethnic groups. Although a case–control study of HIV-infected adults reported greater effectiveness in white people than in black people (20), we know of no data to indicate differences for immunocompetent people. The identical risk for death among black and white people in a recent multivariate analysis of bacteremic pneumococcal pneumonia supports the case-fatality assumption (19). The Appendix describes the all-cause mortality and quality-of-life calculations.
High-Risk Immunocompetent Population
The Advisory Committee on Immunization Practices recommends pneumococcal vaccination for immunocompetent people 2 through 64 years of age with chronic cardiovascular disease, including congestive heart failure and cardiomyopathies; chronic pulmonary disease, including chronic obstructive pulmonary disease and emphysema; diabetes mellitus; alcoholism; chronic liver disease, including cirrhosis; cerebrospinal fluid leaks; or asplenia. Shapiro and colleagues (12) also listed chronic renal failure requiring dialysis in their immunocompetent stratum, although the Advisory Committee on Immunization Practices considers patients with chronic renal failure to be immunocompromised. Combining this information with the comorbid conditions of people with confirmed invasive pneumococcal disease in the CDC database, we included people with congestive heart failure, chronic obstructive pulmonary disease, diabetes mellitus, chronic renal failure, cirrhosis, or chronic alcoholism in the high-risk analysis (Table 1). The Appendix describes the all-cause mortality and quality-of-life calculations.
A recent case–control study of immunocompetent survivors of invasive pneumococcal disease reported that smokers 18 through 64 years of age had 4.1 times the odds of invasive pneumococcal disease, even after multivariate adjustment for age, sex, black race, and chronic illness (21). The addition of smoking to current high-risk factors for the 50- through 64-year-old age group and other age groups warrants future consideration. Following the Advisory Committee on Immunization Practices, we have not included smoking in this analysis.
Low-Risk Population
Case-fatality rates for people 50 through 64 years of age by underlying condition available for 1998 from the CDC's Active Bacterial Core Surveillance Data permitted calculation of health effects for low-risk people separate from those at high risk (5). The 1998 rates for people with no underlying condition—8.8% for black people and 7.6% for nonblack people—were close to the 9.8% in the base-case analysis, while the rates for high-risk people—13.9% for black people and 14.4% for nonblack people—were close to the 19.0% in the best-case analysis. We estimated the immunocompromised group and then derived the low-risk group as those remaining after immunocompromised and high-risk people were subtracted from the total population. The immunocompromised group, consisting of people with HIV infection or AIDS, sickle-cell disease, transplanted organs requiring immunosuppression, or certain types of cancer likely to indicate immunosuppression, accounted for 1.5% of the total population 50 through 64 years of age, 1.86% of nonblack people, and 3.35% of black people (CDC. Unpublished data, 2000; United Network for Organ Sharing, Richmond, Virginia. Unpublished data, 2000; 22). We also incorporated lower disease incidence among low-risk people, estimated from the higher disease incidence among high-risk people and the overall incidence of the general population: 12.5 per 100 000 for the total age group, 27.4 per 100 000 for black people, and 11.0 per 100 000 for nonblack people.
Medical Care Costs
Low-Risk and General Immunocompetent Population
We used Medicare payment rates to approximate the costs of medical care (23). For pneumococcal vaccine and its administration, the base-case amount was the average 1995 Medicare payment and the best- and worst-case amounts were the lowest and highest 1995 Medicare payments, respectively (Table 1). All analyses incorporated the rate of adverse effects after first vaccination for which medical care was sought, 0.11% (1 of 901) (16), and Medicare's 1995 average allowed charge for an office visit for an established patient to approximate treatment cost. Costs of treating invasive pneumococcal disease included only hospitalization. Almost all patients with invasive pneumococcal disease 50 years of age or older are likely to be hospitalized (5), but this conservative assumption excluded related ambulatory costs before or after hospitalization.
High-Risk Populations
Vaccination costs, treatment of adverse vaccination effects, and invasive pneumococcal disease hospitalization for the high-risk group were the same as those for the general population (Table 1). If treatment of high-risk patients with invasive pneumococcal disease was more expensive, hospitalization costs avoided by vaccination would be even greater, as illustrated in the sensitivity analysis.
Role of the Funding Source
The funding source participated in the design of the study and in the decision to submit the manuscript for publication, and provided data on disease incidence, case-fatality rates, and vaccination rates.
Results
|
|---|
|
TopEditors' NotesMethodsResultsDiscussionAuthor & Article InfoReferences |
|---|
One-time vaccination of people 50 through 64 years of age gained an average health benefit of about 1 quality-adjusted day per vaccinee among black people and 0.5 quality-adjusted day among nonblack people (Table 2). Without survivors' future medical costs, vaccination saved about $4 per black person vaccinated and cost $4351 to gain 1 year of healthy life in nonblack people. Since vaccine and its administration cost $15.86, savings in inpatient treatment were about $11 per vaccinee, almost $20 per black person and $10 per nonblack person. Including future medical costs, the cost per QALY gained was $6459 for black people and $12 374 for nonblack people. Estimated adverse effects added $0.03 to vaccination costs.
|
Vaccination cost-effectiveness was generally sensitive to values for invasive pneumococcal disease incidence, vaccination effectiveness, vaccination cost, and inpatient cost. In the global best-case analysis for black people, with all the values most favorable to the vaccination, vaccination, on average, saved about $70 and gained about 3 quality-adjusted days (Table 2). In the global worst-case analysis, with all the values least favorable to vaccination, vaccination cost about $21 500 per QALY gained. Including survivors' future medical costs, the cost-effectiveness ratio remained less than $10 000 per QALY in one-way sensitivity analysis (not shown) and reached $29 168 per QALY in the global worst-case analysis.
Among nonblack people, the results of one-way sensitivity analyses were no higher than $12 155 per QALY excluding future medical costs and $20 112 per QALY including future medical costs (not shown). Global worst-case ratios varied within a narrow range, $68 871 to $76 690 per QALY (Table 2). In probabilistic sensitivity analyses consisting of 1000 Monte Carlo simulations and excluding survivors' medical costs, the 95% probability interval for the cost-effectiveness ratios was narrower than the extremes of the global best- and worst-case ratios. This interval ranged from cost-saving to $1594 per QALY for black people and from cost-saving to $12 273 per QALY for nonblack people.
High-Risk Population
Vaccination generated greater average health benefits among high-risk people, especially black people (Table 2). Excluding future medical costs, vaccination savings ranged from about $6 for nonblack people to $28 for black people. Given vaccination costs, vaccination thus saved about $43 in inpatient treatment for black people and about $22 for nonblack people. Including future medical costs, vaccination cost ranged from $14 721 to $19 128 per QALY for black people and nonblack people, respectively.
In sensitivity analyses for black people, excluding future medical costs, cost savings and health benefits persisted except for the global worst-case analysis (Table 2). Including future medical costs, the highest cost-effectiveness ratio in one-way sensitivity analyses was $18 666 per QALY and the global worst-case ratio was $33 837 per QALY. Among nonblack people, the cost-effectiveness ratio in one-way sensitivity analyses reached $1860 per QALY excluding future medical costs and $23 960 per QALY including future medical costs (not shown). Global best-case ratios ranged from cost savings and health benefits excluding future medical costs to $7550 per QALY including the costs. Global worst-case ratios were $39 000 and $61 565 per QALY, excluding and including future medical costs, respectively.
Low-Risk Population
Compared with the base-case analysis, this model resulted in greater health benefits for high-risk people and general immunocompetent black people and lower health benefits for low-risk people (Table 3). Reflecting changes in assumed disease incidence and hospitalization, savings were higher for high-risk people and costs were higher for low-risk people. The consequent cost-effectiveness ratios for low-risk people ranged from $2477 per QALY for black people to $8195 for nonblack people.
|
Population Estimates
In 1995, 10% of the 34.1 million general population 50 to 64 years of age reported ever having had a pneumococcal ("pneumonia") vaccination (8). Among the 6.8 million people with certain chronic conditions, the rate was 20.1% overall and 11.7% among black people. Although these conditions are broader than the high-risk comorbid conditions in our analysis and exclude alcoholism, this rate is the best available proxy for people at high risk for pneumococcal disease. By 2001, these rates had increased to only 15.4% for the general population, 26.7% overall for those with chronic conditions, and 12.5% for black people with chronic conditions (CDC. Unpublished data, 2003).
The magnitude of total health effects and costs to be gained depend on the vaccination rates that can be achieved. For the general immunocompetent population 50 through 64 years of age who were unvaccinated in 1995, with base-case results, the following illustrate potential implications:
1. Vaccination rates increased to 32% (the annual influenza vaccination rate for this age group in 2001) would have gained 11 102 QALYs and cost $38.1 million excluding future medical costs and $126.7 million including future medical costs.
2. Vaccination rates increased to 52% (the pneumococcal vaccination rate for people 65 years of age or older in early 2001) would have gained 21 195 QALYS and cost $72.8 million excluding future medical costs and $242 million including future medical costs.
3. Vaccination rates increased to 90% (the objective for people 65 years of age or older to be reached by 2010) would have gained 40 327 QALYs and cost $138.6 million excluding future medical costs and $460.9 million including future medical costs (24) (CDC. Unpublished data, 2003).
Savings in invasive pneumococcal disease hospitalizations would have totaled $81.1 million for 32%, $154.9 million for 52%, and $295.0 million for 90% rates.
For high-risk people, increasing vaccination rates to 41% (the annual influenza vaccination rate for those in the 50- through 64-year-old age group in 2001) would have gained 2860 QALYs and saved $12.1 million excluding future medical costs or cost $51.8 million including future medical costs; increasing the rate to 60% (the 2010 objective for pneumococcal vaccination among noninstitutionalized high-risk adults) would have gained 5460 QALYs and saved $23.2 million excluding future medical costs or cost $98.9 million including future medical costs (24) (CDC. Unpublished data, 2003). Preventing invasive pneumococcal disease hospitalization would have saved $34.7 million for 41% and $66.3 million for 60% rates. For low-risk people, estimated to number 27.4 million in 1995, raising vaccination rates from the estimated 8% in 1995 to 28% (the 2001 annual influenza vaccination rate for people without high-risk conditions in this age group) would have gained 6153 QALYS, cost $46.5 million excluding future medical costs and $69.3 million including future medical costs, and saved $40.7 million in invasive pneumococcal disease hospitalization prevented (CDC. Unpublished data, 2003).
Discussion
|
|---|
|
TopEditors' NotesMethodsResultsDiscussionAuthor & Article InfoReferences |
|---|
Such underuse characterizes many proven effective, cost-effective interventions and is associated with racial and ethnic disparities in health-related outcomes (25). For example, influenza vaccination rates for high-risk people 50 through 64 years of age in 1995 stood at 40.4% for non-Hispanic white people but only 24.5% for non-Hispanic black people (8). Similar relationships have been found for other preventive and therapeutic services, such as mammograms and analgesia for metastatic cancer (25).
Vaccination may also be a wise investment for non–high-risk immunocompetent people 50 through 64 years of age, especially for black people. Incremental cost to gain 1 year of healthy life—about $2500 for low-risk black people, about $8200 for nonblack people, and about $3400 for the age group overall—is lower than that of other preventive measures now recommended for people 50 to 64 years of age. In a 1996 analysis, screening people from age 50 years for colorectal cancer cost $8000 to $18 000 per life-year gained, depending on the technology (26). A more recent analysis of screening from age 50 years with 1998 costs reported similar results: incremental cost-effectiveness of $9705 for annual fecal occult blood testing and $10 983 for colonoscopy every 10 years (27). Although the authors did not adjust for quality of life, their methods were otherwise similar to those used in this analysis. For pneumococcal vaccination, even cost-effectiveness ratios including future medical costs (about $19 000 for nonblack high-risk people) lie within the range usually considered reasonable to gain 1 year of healthy life (28).
These analyses have incorporated conservative estimates of health benefits and cost-savings from vaccination. Based on the period covered by the case–control study, the analyses assumed that protection against disease lasts 6 years (12). Preventing disease would also avoid some ambulatory costs, which this analysis did not incorporate. The analysis also excluded any benefits from prevention of pneumococcal pneumonia, separate from invasive pneumococcal disease. On the other hand, implementing recommendations to increase existing vaccination rates may require additional costs, for example, to undertake standing orders for nurses or computer-based reminders for clinicians. Vaccination would continue to be cost-saving for high-risk people for implementation costs up to about $8 per vaccinee (Table 2).
Although the results for vaccinating people 50 through 64 years of age are generally favorable, those for universal vaccination for people 65 years of age or older are even more compelling (3). With a similar model, vaccination was cost-saving for people 65 years of age and older and for each of the subgroups: 65 through 74 years, 75 through 84 years, and 85 years and older. Given the epidemiology of disease, universal adult vaccination at ages younger than 50 years is unlikely to be as cost-effective.
Since cost-effectiveness analysis, unlike cost–benefit analysis, expresses health effects in natural units related to morbidity and mortality, we did not convert these effects into monetary terms. Consistent with the recommendations of the federal Panel on Cost-Effectiveness in Health and Medicine, we included the effects on morbidity along with mortality in the denominator, as more QALYs (10). The QALY measure thus incorporates the full effect of vaccination on morbidity and mortality (10). This better health could lead to fewer absences and better productivity at work, which the Panel termed "time costs" and others have termed "indirect costs." In our cost-effectiveness analysis, we used QALYs to incorporate the potential implications of better health, including such differences in productivity, but did not array them separately. By contrast, a cost–benefit analysis would incorporate the financial implications of these health effects, such as better productivity.
This cost-effectiveness analysis did not address the implications of pneumococcal revaccination. The risk for adverse effects seems to be low for one-time revaccination (16, 29). Concern centers on whether non–high-risk people initially vaccinated before the currently recommended age of 65 years will have less protection from revaccination after age 65 years, when the incidence of pneumococcal disease and associated mortality are higher. There are no data on the clinical effectiveness of revaccination, although serologic studies suggest that immune responses are lower after a second dose of pneumococcal polysaccharide vaccine than after the first dose (6, 7). If subsequent doses are less effective, revaccination for people 65 years of age or older might achieve fewer overall health benefits and be less cost-effective than primary vaccination at that age, which an earlier study found is likely to be cost-saving (3). The clinical and economic implications of increasing antibiotic resistance in pneumococcal infection also deserve consideration. Given the lack of relevant evidence, research on these issues is clearly warranted.
These cost-effectiveness results can inform the deliberations of the Advisory Committee on Immunization Practices and other organizations considering recommendations for a general evaluation at age 50 years. They confirm the merit of making implementation of the current recommendation for vaccinating high-risk people in this age group a high priority. The findings suggest that expanding the recommendation for pneumococcal vaccination to the general population 50 to 64 years, especially for black people, may be cost-effective. Since disease risk increases with age, the findings also highlight the importance of addressing the effectiveness of revaccination for people 65 years of age or older and the current epidemiology of pneumococcal disease in formulating vaccination policy.
Appendix
|
|---|
We derived the effectiveness of vaccination to prevent invasive pneumococcal disease caused by vaccine serotypes from results calculated by a case–control study (12). That study reported point estimates of vaccination effectiveness and 95% confidence limits at 0 through 2 years, 3 through 5 years, and more than 5 years after vaccination by age (<55 years, 55 through 64 years, and 65 through 74 years) (12). To estimate effectiveness by year for our base-case analysis, we used simple regression (ordinary least-squares) to relate the reported point estimate for vaccine effectiveness to the number of years since vaccination. For the best- and worst-case analyses, we used simple regression models that correlated the reported upper and lower 95% confidence limits with time since vaccination. We then applied the age-specific estimates to derive estimates for people of specific ages in our hypothetical cohort of the U.S. population 50 through 64 years of age.
General Immunocompetent Population
All-Cause Mortality
The analysis used secular trends in all-cause mortality to reflect the probability of death among those who did and did not contract invasive pneumococcal disease (Table 1). By using 1992 all-cause mortality rates and projections for 2000 and 2050, we calculated age-weighted mortality rates for 10-year age groups (U.S. Census Bureau. Unpublished data, 1998). We estimated linear time trends for each age group and then annual mortality rates until the youngest reached 100 years of age. For each year, we also derived mortality rates from other causes for those who did and did not contract invasive pneumococcal disease, based on the theory of competing risks (30).
Quality of Life
We used years-of-healthy-life measures based on the 1990 National Health Interview Survey (NHIS) to adjust expected years of life to reflect that survivors would experience less than perfect health (Table 1) (14, 15). We assumed that quality of life for the year for people without invasive pneumococcal disease equaled the average health-related quality of life by age. We drew the weights for survivors' quality of life from calculations of years of healthy life that Torrance, Erickson, and colleagues developed from age-specific responses to activity limitation and perceived health status in the National Health Interview Survey (14, 15). They defined six levels of activity limitation from questions about a person's ability to perform his or her usual social role (not limited, limited in other activities, limited in major activity, unable to perform major activity, limited in instrumental activities of daily living, or limited in activities of daily living) and five categories of perceived health status from the question on a person's rating of his or her health status (excellent, very good, good, fair, or poor).
The two concepts of activity limitation and perceived health status jointly defined 30 living health states plus the dead state. With 1 as the best health state (activity not limited and perceived health excellent) and 0 as dead or the worst, they then used multiattribute utility scaling and correspondence analysis to assign scores to each of the activity limitation and perceived health status attributes. These scores, along with a corner value from the Health Utilities Index Mark I, enabled computation of the remaining pairs of activity limitation and health status attributes. Like ratings of health status in the Medical Outcomes Study, the results from Erickson and colleagues (14) and Torrance and colleagues (15) on the perceived health status scale were nonlinear. For years of life among survivors in the general immunocompetent population, we used the average scores by age group (14, 15). The score for people 50 through 55 years of age, 0.83, was closest to the value for someone in the National Health Interview Survey who reported good health status and no activity limitations, 0.84.
On the basis of bacteremia, we assumed that invasive pneumococcal disease patients had 34 days of restricted-activity or bed days, with limited performance of activities of daily living and "fair" perceived health status, a weight of 0.20 (Table 1) (National Center for Health Statistics, National Health Interview Survey, condition file, unspecified septicemia. Unpublished data, 1993). These restricted activity days pertained to all locations, including hospital and home. The quality of life for 1 year for a patient with invasive pneumococcal disease was the weighted sum of the quality of life with invasive pneumococcal disease and the average age-specific quality of life, according to the number of days of the year spent with and without the disease. If convalescence in fact lasts longer than our assumption, the health benefits of prevention would be greater.
High-Risk Immunocompetent Population
All-Cause Mortality
We chose heart failure to exemplify likely all-cause mortality and medical costs associated with a high-risk condition among survivors of invasive pneumococcal disease.
Given our modeling framework and components, we required a high-risk condition for which there were data on mortality rates from all causes from age 50 years through life expectancy, preferably from a nationally representative sample. Schocken and colleagues (13) had combined data on the prevalence of heart failure from the federal National Health and Nutrition Examination Survey (NHANES I) with data on mortality rates from the NHANES I Epidemiology Follow-up Study. We incorporated NHANES data linked to derive mortality rates through 74 years of age (13), applied these rates through age 79 years, and assumed that thereafter, the excess mortality of this high-risk cohort would continue to be the same multiple of the general population's mortality rate.
Quality of Life
We drew the weights for quality of life among high-risk people, as we had for the general immunocompetent population, from the calculations by Torrance and colleagues and Erickson and colleagues (Table 1) (14, 15).
For a person in the high-risk group, defined as immunocompetent people with congestive heart failure, chronic obstructive pulmonary disease, diabetes mellitus, chronic renal failure, cirrhosis, or chronic alcoholism, a lower-scoring category in terms of activity limitation or health status seemed appropriate. The category of limited performance of activities other than major activities and a "good" perceived health status, a quality-of-life value of 0.72 versus 0.83 for this age group in the general population, seemed most reasonable for high-risk people and a conservative assumption (Table 1) (14, 15). If a high-risk person's health was poorer and closer to the next category of fair health status and no limitation, 0.63, the health benefits of preventing invasive pneumococcal disease would be even greater.
For survivors' quality of life, we assumed that from 50 through 55 years of age, high-risk people would have a quality-of-life value of 0.72. We then assumed that the health status of high-risk people declined with age, so that by 75 through 79 years of age, it was the same as that of the general population 85 years of age or older (Table 1). The quality of life for high-risk people who developed invasive pneumococcal disease was assumed to be associated with limited performance of activities of daily living and "fair" perceived health status during the illness, 0.20, regardless of age. If the quality of life of patients with invasive pneumococcal disease decreased with advancing age or comorbid conditions, the health benefits from preventing disease would be even greater.
To illustrate the implications of higher and lower QALY weights, the sensitivity analyses varied each weight for average health plus and – 0.05 for the best and worst cases, respectively, and each weight for the invasive pneumococcal disease state minus and plus 0.05 for the best and worst cases, respectively (Table 1).
Author and Article Information
|
|---|
|
TopEditors' NotesMethodsResultsDiscussionAuthor & Article InfoReferences |
|---|
Acknowledgments: The authors thank the Centers for Disease Control and Prevention staff, especially Ray Strikas, Walter Williams, and Walter Orenstein, for encouragement and guidance, and James Singleton for data on immunocompromised people; the United Network for Organ Sharing for data on transplant survivors; and Marshall McBean, University of Minnesota, and the Centers for Medicare & Medicaid Services (formerly the Health Care Financing Administration) for data on Medicare payment rates.
Grant Support: This project was supported under a cooperative agreement from the Centers for Disease Control and Prevention through the Association of Schools of Public Health, project number S343-16/16.
Potential Financial Conflicts of Interest: None disclosed.
Requests for Single Reprints: Jane E. Sisk, PhD, Department of Health Policy, Mount Sinai School of Medicine, Room 2-34, 1425 Madison Avenue, New York, NY 10029; e-mail, jane.sisk{at}mssm.edu.
Current Author Addresses: Dr. Sisk: Department of Health Policy, Mount Sinai School of Medicine, Room 2-34, 1425 Madison Avenue, New York, NY 10029.
Dr. Whang: Department of Medicine, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114.
Dr. Butler: Centers for Disease Control and Prevention, 4055 Tudor Centre Drive, Anchorage, AK 99508.
Dr. Sneller: Centers for Disease Control and Prevention, 1600 Clifton Road, MS C-61, Atlanta, GA 30333.
Dr. Whitney: Centers for Disease Control and Prevention, 1600 Clifton Road, MS C23, Atlanta, GA 30333.
Author Contributions: Conception and design: J.E. Sisk, J.C. Butler, V.-P. Sneller.
Analysis and interpretation of the data: J.E. Sisk, W. Whang, J.C. Butler, V.-P. Sneller, C.G. Whitney.
Drafting of the article: J.E. Sisk.
Critical revision of the article for important intellectual content: J.E. Sisk, W. Whang, J.C. Butler, V.-P. Sneller, C.G. Whitney.
Final approval of the article: J.E. Sisk, W. Whang, J.C. Butler, V.-P. Sneller, C.G. Whitney.
Provision of study materials or patients: J.C. Butler, V.-P. Sneller, C.G. Whitney.
Statistical expertise: W. Whang.
Obtaining of funding: J.E. Sisk, V.-P. Sneller.
Administrative, technical, or logistic support: J.E. Sisk, J.C. Butler, V.-P. Sneller.
Collection and assembly of data: J.E. Sisk, W. Whang, C.G. Whitney.
References
|
|---|
|
TopEditors' NotesMethodsResultsDiscussionAuthor & Article InfoReferences |
|---|
1. Prevention of pneumococcal disease: recommendations of the Advisory Committee on Immunization Practices (ACIP) MMWR Recomm Rep. 1997;46:1-24. [PMID: 9132580].[Medline]
2. U.S. Preventive Services Task Force. Guide to Clinical Preventive Services. 2nd ed. Baltimore: Williams & Wilkins; 1996.
3. Sisk JE, Moskowitz AJ, Whang W, Lin JD, Fedson DS, McBean AM, et al. Cost-effectiveness of vaccination against pneumococcal bacteremia among elderly people JAMA. 1997;278:1333-9. [PMID: 9343464].
4. Pneumococcal and influenza vaccination levels among adults aged > or = 65 years—United States, 1995 MMWR Morb Mortal Wkly Rep. 1997;46:913-9. [PMID: 9347904].[Medline]
5. Robinson KA, Baughman W, Rothrock G, Barrett NL, Pass M, Lexau C, et al. Epidemiology of invasive Streptococcus pneumoniae infections in the United States, 1995–1998: Opportunities for prevention in the conjugate vaccine era JAMA. 2001;285:1729-35. [PMID: 11277827].
6. Borgoño JM, McLean AA, Vella PP, Woodhour AF, Canepa I, Davidson WL, et al. Vaccination and revaccination with polyvalent pneumococcal polysaccharide vaccines in adults and infants Proc Soc Exp Biol Med. 1978;157:148-54. [PMID: 23549].[Medline]
7. Mufson MA, Hughey DF, Turner CE, Schiffman G. Revaccination with pneumococcal vaccine of elderly persons 6 years after primary vaccination Vaccine. 1991;9:403-7. [PMID: 1887670].[Medline]
8. Singleton JA, Greby SM, Wooten KG, Walker FJ, Strikas R. Influenza, pneumococcal, and tetanus toxioid vaccination of adults—United States, 1993–7 MMWR CDC Surveill Summ. 2000;49:39-62. [PMID: 11016877].[Medline]
9. Influenza and pneumococcal vaccination levels among persons aged > or = 65 years—United States, 1999 MMWR Morb Mortal Wkly Rep. 2001;50:532-7. [PMID: 11446571].[Medline]
10. Gold MR, Siegel JE, Russell LB, Weinstein MC. Cost-Effectiveness in Health and Medicine. New York: Oxford Univ Pr; 1996.
11. Weinstein MC, Stason WB. Foundations of cost-effectiveness analysis for health and medical practices N Engl J Med. 1977;296:716-21. [PMID: 402576].[Abstract]
12. Shapiro ED, Berg AT, Austrian R, Schroeder D, Parcells V, Margolis A, et al. The protective efficacy of polyvalent pneumococcal polysaccharide vaccine N Engl J Med. 1991;325:1453-60. [PMID: 1944423].[Abstract]
13. Schocken DD, Arrieta MI, Leaverton PE, Ross EA. Prevalence and mortality rate of congestive heart failure in the United States J Am Coll Cardiol. 1992;20:301-6. [PMID: 1634664].[Abstract]
14. Erickson P, Wilson R, Shannon I. Statistical notes no. 7: Years of healthy life. In: Healthy People 2000. Atlanta: Centers for Disease Control and Prevention, National Center for Health Statistics; April 1995.
15. Torrance GT, Erickson P, Patrick DL, Feldman JJ. Technical notes. In: Erickson P, Wilson R, Shannon I. Statistical notes no. 7: Years of healthy life. In: Healthy People 2000. Atlanta: Centers for Disease Control and Prevention, National Center for Health Statistics; April 1995.
16. Jackson LA, Benson P, Sneller VP, Butler JC, Thompson RS, Chen RT, et al. Safety of revaccination with pneumococcal polysaccharide vaccine JAMA. 1999;281:243-8. [PMID: 9918479].
17. Agency for Health Care Policy and Research. Data from the 1987 NMES Household Survey Projected to 1995–2005: Medical Expenditures by Type of Service and Payment Source, Demographics, and Poverty and Insurance Status. Rockville, MD: Agency for Health Care Policy and Research; December 1997.
18. Laschober MA. Health and Health Care of the Medicare Population: Data from the 1993 Medicare Current Beneficiary Survey. Rockville, MD: Westat; 1997.
19. Feikin DR, Schuchat A, Kolczak M, Barrett NL, Harrison LH, Lefkowitz L, et al. Mortality from invasive pneumococcal pneumonia in the era of antibiotic resistance, 1995–1997 Am J Public Health. 2000;90:223-9. [PMID: 10667183].
20. Breiman RF, Keller DW, Phelan MA, Sniadack DH, Stephens DS, Rimland D, et al. Evaluation of effectiveness of the 23-valent pneumococcal capsular polysaccharide vaccine for HIV-infected patients Arch Intern Med. 2000;160:2633-8. [PMID: 10999977].
21. Nuorti JP, Butler JC, Farley MM, Harrison LH, McGeer A, Kolczak MS, et al. Cigarette smoking and invasive pneumococcal disease. Active Bacterial Core Surveillance Team N Engl J Med. 2000;342:681-9. [PMID: 10706897].
22. Winter WP. Hemoglobin Variants in Human Populations. Boca Raton, FL: CRC Pr; 1986.
23. Lave JR, Pashos CL, Anderson GF, Brailer D, Bubolz T, Conrad D, et al. Costing medical care: using Medicare administrative data Med Care. 1994;32:JS77-89. [PMID: 8028415].[Medline]
24. Objectives for Improving Health. In: Healthy People 2010. 2nd ed. vol 1. Atlanta: Centers for Disease Control and Prevention; 2001:14-45. Accessed at http://www.healthypeople.gov/document on 27 February 2003.
25. Fiscella K, Franks P, Gold MR, Clancy CM. Inequality in quality: addressing socioeconomic, racial, and ethnic disparities in health care JAMA. 2000;283:2579-84. [PMID: 10815125].
26. Wagner JL, Tunis S, Brown M, Ching A, Almeida R. The cost effectiveness of colorectal cancer screening in average-risk adults. In: GP Young, P Rosen, B Levin, eds. Prevention and Early Detection of Colorectal Cancer. Philadelphia: WB Saunders; 1996.
27. Sonnenberg A, Delcò F, Inadomi JM. Cost-effectiveness of colonoscopy in screening for colorectal cancer Ann Intern Med. 2000;133:573-84. [PMID: 11033584].
28. Laupacis A, Feeny D, Detsky AS, Tugwell PX. Tentative guidelines for using clinical and economic evaluations revisited Can Med Assoc J. 1993;148:927-9. [PMID: 11651619].[Medline]
29. Nichol KL, MacDonald R, Hauge M. Side effects associated with pneumococcal vaccination Am J Infect Control. 1997;25:223-8. [PMID: 9202818].[Medline]
30. Chiang CL. Competing risks in mortality analysis Annu Rev Public Health. 1991;12:281-307. [PMID: 2049139].[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
Related articles in Annals:
This article has been cited by other articles:
|
|
 |
|
  L. B. Russell Preventing Chronic Disease: An Important Investment, But Don't Count On Cost Savings Health Aff., January 1, 2009; 28(1): 42 - 45. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. Neumann, P. D. Jacobson, and J. A. Palmer Measuring the Value of Public Health Systems: The Disconnect Between Health Economists and Public Health Practitioners Am J Public Health, December 1, 2008; 98(12): 2173 - 2180. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. C. Meyer Aging Proceedings of the ATS, December 1, 2005; 2(5): 433 - 439. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. R. Sokos Pharmacists' role in increasing pneumococcal and influenza vaccination Am. J. Health Syst. Pharm., February 15, 2005; 62(4): 367 - 377. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Flannery, S. Schrag, N. M. Bennett, R. Lynfield, L. H. Harrison, A. Reingold, P. R. Cieslak, J. Hadler, M. M. Farley, R. R. Facklam, et al. Impact of Childhood Vaccination on Racial Disparities in Invasive Streptococcus pneumoniae Infections JAMA, May 12, 2004; 291(18): 2197 - 2203. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. A. McGlynn There Is No Perfect Health System Health Aff., May 1, 2004; 23(3): 100 - 102. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Fetterolf and R. West The Business Case for Quality: Combining Medical Literature Research with Health Plan Data to Establish Value for Nonclinical Managers American Journal of Medical Quality, March 1, 2004; 19(2): 48 - 55. [Abstract] [PDF]
|
|
|
|
 |
|
 When to Give Pneumococcal Vaccine -- Younger May Be Better Journal Watch Infectious Diseases, September 5, 2003; 2003(905): 1 - 1. [Full Text]
|
|
|
|
 |
|
 Pneumococcal Vaccination Starting at Age 50? Journal Watch (General), July 18, 2003; 2003(718): 2 - 2. [Full Text]
|
|
|
|
 |
|
 P. Gardner A Need To Update and Revise the Pneumococcal Vaccine Recommendations for Adults Ann Intern Med, June 17, 2003; 138(12): 999 - 1000. [Full Text] [PDF]
|
|
Article
