In the past decade, reported pertussis incidence has increased by 106% in individuals 10–19 years of age and by 93% in those older than the age of 20 years, with large outbreaks occurring every 3–4 years.1 Despite accepted case definitions, significant underreporting exists, particularly in adolescents and adults.2
An acellular formulation of the pertussis vaccine, more tolerable than the whole cell formulation, was recommended in 1997 for routine use in American infants and children. Currently an acellular pertussis vaccine is administered at 2, 4 and 6 months of age, with a booster dose given at 15–18 months and 4–6 years of age. Expanding immunization beyond infancy and childhood to include adolescents could lead to better control of pertussis and could be especially beneficial because adolescents can be a significant source of infection to infants,3–5 in whom the risk of morbidity and mortality from pertussis is highest.
The costs of pertussis vary by age group. A recent study in the United States6 estimated the direct medical costs per case to be $2822 for infants, $308 for children, $254 for adolescents and $181 for adults (year 2000 US dollars). Hospitalizations increased the cost per infant infected, whereas in adolescents and adults, physician visits contributed most to the total cost. Thus prevention of significant numbers of infant cases may offset the costs of expanded immunization, whereas the direct effects in the vaccinated group(s) are unlikely to do so alone.
Despite intensive efforts in epidemiologic modeling of pertussis, economic analysis of the impact of pertussis immunization has been limited, especially in terms of the scope of questions being addressed. Determining the economic efficiency of a pertussis immunization strategy not only is a question of whether to vaccinate or not but also depends on who is to be vaccinated, how to implement the program and which type of vaccine to use.7 Economic models addressing some of these aspects have been published for the United States,8–11 Canada,12,13 several countries in Europe14–18 and one in Australia.19 Some of the models, however, are quite old, and the data used are no longer current. Moreover, some of the strategies modeled, such as whole cell vaccination versus no vaccination, are now outdated.
The objective of this study was to formally evaluate, using an up-to-date model, the economic outcomes of expanding pertussis immunization to adolescents.
The model used in these analyses estimates the health economic outcomes of adding an adolescent acellular booster dose (between the ages of 11 and 18 years) to the current U.S. pertussis immunization schedule. It is assumed that the acellular pertussis vaccine is given as a combined pertussis-diphtheria-tetanus (dTaP) vaccine, replacing the current immunization practice of vaccinating with a combined diphtheria-tetanus (dT) vaccine. A single cohort of potential vaccinees is followed during their lifetime (maximum, 105 years). The incidence and consequences of pertussis in the rest of the population are tracked for the same length of time.
The model starts by replicating the current epidemiology of pertussis in the United States, using age-specific epidemiologic and cost data to create the “current immunization schedule” scenario. The population is categorized into 4 broad age groups for reporting and for some of the inputs: infants (younger than 1 year old), children (1–10 years old), adolescents (11–18 years old) and adults (older than 18 years). Within the model, the cohorts are modeled for each year of age. The addition of an adolescent booster dose is then modeled in 2 distinct modules: one addresses the effects in the target subgroup; and the other considers the indirect effects in other age groups.
The impact on vaccinated individuals is represented by the reduction in pertussis incidence in this group, based on known vaccine efficacy. Waning vaccine-induced immunity and modification of the severity of breakthrough pertussis cases are also taken into consideration.
The indirect impact on other age groups and on unvaccinated adolescents is modeled by assuming that infection will be decreased because of reduced transmission of the disease resulting from the “herd” immunity effect of adding an adolescent booster vaccine. The size of the potential herd effect, its persistence over time and the relative impact on different age groups were all taken into consideration. Like the efficacy of the vaccine and its waning effect, the size, persistence and age-related impact of herd immunity are all inputs into the model; they are not calculated by the model.
Outcomes are calculated from both the health care payer perspective (direct costs only) and the societal perspective (direct and indirect costs) and are discounted at 3% per annum after the first year.20
Data for the model were sought from a variety of sources, including government reports and statistics, administrative data sets and the published literature. Because review of publicly available data did not yield all the required inputs, estimates for these were obtained from members of the Global Pertussis Initiative (GPI), an expert forum of 37 members established in 2001 to address the neglected problem of pertussis. As part of this Initiative, experts were provided with a table containing a description of the parameters in question and a suggested value and range based on the literature review. Members of the GPI were then asked whether they agreed with the range; if they disagreed, they were asked to provide alternative estimates for the parameters. Table 1, which outlines the main inputs, indicates which values were based on responses from the GPI members.
The age-specific incidence of reported cases of pertussis was calculated and was based on the average of estimates from the Centers for Disease Control and Prevention (CDC) from 1996 to 1999. Cases reported by the CDC ranged from 6549 in 1997 to 7771 in 1996. Pertussis incidence was highest in infants, at just under 57 cases per 100,000 person-years, and decreased with age, falling to <1 case per 100,000 person-years for those 25 years of age and above.21–24 It is well-established that underreporting of pertussis is extensive, often because of misdiagnosis or because patients do not present to medical professionals.25–32 Estimates of the extent of underreporting vary substantially, depending on methodology and case definition. In a prospective study based on laboratory testing of patients presenting with persistent cough, the annual incidence of pertussis in adolescents and adults was estimated at 507 per 100,000 person-years,30 suggesting that <1% of cases of pertussis are reported to the CDC. A recent simulation of pertussis in the United States proposed that the annual incidence of pertussis could be >1750 per 100,000 population, with 21.5% being typical cases, 28.2% being mild and just over one-half being asymptomatic.31 Another study, using hospitalization data combined with other published data on pertussis incidence in the United States, estimated a reporting rate of 11.6%.32 Estimates by members of the GPI also varied widely, but the majority agreed on a conservative approach, with adoption of the reporting rate of 11.6% (7.6 unreported cases per reported case).32 It was assumed that 70% of these unreported cases would be significantly milder than typical cases. For typical cases of pertussis, age-specific rates of hospitalization and death were based on data from the CDC.33 For rates of long term disability, the proportion of reported cases of pertussis leading to encephalitis was used,33 and it was assumed that 33% of encephalitis cases would lead to permanent sequelae in survivors.8
Available data on the efficacy of the acellular vaccine in adolescents and adults34,35 are in line with estimates of vaccine efficacy in children.36,37 Thus an initial vaccine efficacy of 85% (ie, the vaccine leads to complete protection in 85% of those vaccinated) was used. Because the data also indicate that immunization modifies severity, an estimated 45% of breakthrough cases were assumed to be significantly milder than the typical disease. Vaccine-induced protection was estimated to wane, with all protection ending 10 years after vaccination. In the main analyses, a linear rate of waning was assumed.
Coverage of 80% of adolescents was assumed. Although the acellular vaccine has been shown to be very safe, it was assumed that 2% of vaccinees would have adverse reactions of sufficient severity to warrant a physician visit. Finally in the base case, a vaccine wastage rate of 10% was applied.
There are no adequate data with which to estimate herd immunity resulting from adolescent pertussis booster immunization. Although adolescent booster immunization has been implemented in France and Germany, estimates of the indirect benefits are not yet available. Moreover efforts at deriving herd immunity via modeling have not provided definitive results.17,38 Members of the GPI estimated that 20% of cases in those not vaccinated would be prevented with implementation of an adolescent booster (range, 5–35%) and that 20% of these avoided cases would have been in infants, 30% in children, 25% in unvaccinated adolescents and 25% in adults. The duration of herd immunity was assumed to be consonant with the protection conferred upon vaccinees.
The costs used in this analysis are estimated to reflect the economic value of the resources consumed, regardless of who pays for them. Therefore they do not take into account adjustments made in the multipayer U.S. system, such as copayments, deductibles or volume discounts. Cost profiles for typical cases of pertussis were developed for each age group, covering physician visits, emergency room visits, hospitalizations, long term disability, antibiotics and other drugs, diagnostic tests and prophylactic measures. The indirect costs related to pertussis addressed lost time from paid work activities, which could be as a result of the illness itself, caring for someone with pertussis or lack of employability because of serious sequelae from pertussis disease. Time lost from nonwork activities was not included. Mild breakthrough or unreported cases were assumed not to lead to hospitalization or long term disability and to result in direct and indirect costs that are 45% lower than those relating to typical cases who neither require hospitalization nor experience long term disability.
All cost estimates are reported in 2002 US dollars. Where 2002 values were not available, older estimates were inflated using the Medical Care Inflation Index (a component of the U.S. Consumer Price Index) supplied by the Federal Bureau of Labor Statistics for January of each relevant year. Any charges used as inputs were adjusted to costs using a cost-to-charge ratio of 0.61 provided by the Commonwealth of Massachusetts Office of Health Care Finance and Policy. Costs were adjusted to reflect national values using ratios based on geographical variations published by the Centers for Medicare and Medicaid Services.39–56
The cost of adding acellular pertussis vaccine to dT vaccine was estimated by subtracting the cost of the dT vaccine from that of the acellular vaccine. After addition of wastage and the treatment of adverse reactions, the net cost was estimated at $18.15 per vaccination. Age-specific costs per case of pertussis for diagnostic tests ($32.75–$47.06), antibiotics ($11.65–$22.35), other drug use ($0–$8.24) and prophylaxis ($16.76) were based on a U.S. study,6 as were the number of physician (1.5–4.5) and emergency room visits (0–0.36). Costs of these visits came from a statewide emergency room encounter database and fee schedules46,49–51 ($38.48–$41.76 for physician visits and $170.24 for emergency room visits). Costs per hospitalized case were based on data from five 1999 all-payer state inpatient discharge databases41–45 and ranged from $5500 in adolescents and adults to almost $10,000 in infants. Average direct costs per typical case, incorporating all of the above costs, are presented in Table 1. Time lost from paid work activities was taken from a U.S. study,6 supplemented by information on adolescents and adults published in a recent Canadian analysis.57 Average U.S. hourly wages ($17.29)53 were used to value this time. The cost of time lost from work in infected adults was estimated at $607 per case, whereas costs of caring for individuals with pertussis ranged from $22 to $330, depending on the age of the patient. Lost wages for individuals with severe long term sequelae were calculated using U.S. employment and income statistics.54–56
Total direct and indirect costs with and without immunization were calculated, along with the number of pertussis cases, the number of individuals suffering permanent sequelae, and pertussis-related deaths and life-years lost. Based on these results, cost-effectiveness ratios for immunization were calculated as the cost per case avoided and the cost per life-year gained (LYG). Given the uncertainty in parameter estimates, the results of a large number of sensitivity analyses are presented. Because adequate data on herd immunity are lacking, broad ranges were considered. Because infant cases of pertussis are the most serious, we varied the proportion of the herd immunity effect in infants from 5% to 35% and examined the extreme scenario where all cases avoided because of herd immunity are in infants. Waning of immunity was also subjected to functional forms other than linear.
With a 20% reduction in pertussis cases among those not vaccinated, extending immunization to a single cohort of adolescents would, in the subsequent 10 years, prevent 68,408 cases of pertussis in the United States, including 41 deaths, which translates to >3000 LYGs (Table 2). Immunization of 80% of the target adolescent cohort would cost nearly $75 million but would result in a reduction of almost $50 million in direct costs and a further $20 million in indirect costs. These outcomes yield a cost effectiveness ratio of about $6000 per LYG when all costs are included, and $22,000 per LYG when only medical costs are included. At a 35% reduction in cases among those not vaccinated, adolescent immunization becomes the dominant strategy (ie, produces health benefits and lowers costs), reducing pertussis cases by >100,000 and leading to savings in tens of millions of dollars. Conversely if there were only a 5% reductionin other cases, the cost effectiveness deteriorates above $150,000 per discounted LYG.
Figure 1 extends the analysis across the entire range of herd immunity, keeping all other model inputs constant. Pertussis immunization of adolescents becomes cost-neutral from the health care perspective if herd immunity is slightly more than 30%. A cost per LYG of <$20,000 is attained at herd immunity levels of between 16 and 21%; it remains below $50,000 per LYG if herd immunity is at least 12%. From the societal perspective, adolescent immunization is dominant if there is a 25% indirect reduction of cases. If there were no herd immunity at all, a very unlikely result, fewer than 5000 cases of pertussis would be prevented, and the cost per discounted case avoided would rise to >$16,000.
Table 3 presents results varying a number of other key inputs. Given the importance of herd immunity, these are presented for the base case of 20% and for the lower and upper boundaries of 5 and 35%. At the base case of 20%, changes in inputs can play a decisive role, although in most cases, cost effectiveness ratios are quite positive. If there are more unreported cases than the 7 estimated in the base case or if fewer unreported cases are mild, adolescent immunization becomes dominant. Conversely if the vaccine is only effective for 3 years or if adding the acellular pertussis vaccine to the existing dT increased administration costs by $20, then cost effectiveness deteriorates considerably. Varying vaccine effectiveness has a very weak effect on cost effectiveness outcomes, partly because changes in vaccine efficacy are affecting only vaccinated individuals, whereas the herd immunity effect is held constant.
The full effect of underreporting is shown in Figure 2, demonstrating that over wide ranges, the true incidence of pertussis is equally as important as herd immunity in determining cost effectiveness. In fact, if underreporting is extensive and these unreported cases are symptomatic, then herd immunity levels become less critical to the economic evaluation. When recently published age-specific incidence estimates31 were incorporated into the model (380 per 100,000 for typical cases, and 498 per 100,000 for mild cases for all ages combined), results altered substantially. Even in the absence of any herd immunity, vaccination would lead to overall savings in direct costs of almost $19 million and total savings of almost $56 million. Without herd immunity, at these rates of infection, costs per case could be between 30 and 45% lower before savings would be eliminated. With 20% herd immunity, costs per case could be >90% lower.
The results of these analyses suggest that adding an adolescent pertussis booster dose to the current infant and toddler immunization schedules in the United States would be cost-effective, given reasonable estimates of herd immunity, and applying the assumption that just under 12% of all pertussis cases are reported to the CDC. Regardless of this, adequate coverage must be attained; this is best done by coupling the acellular pertussis vaccine with the existing dT booster. This also keeps the administration costs to a minimum.
These estimates of the economic and health implications of adolescent immunization in the United States must be viewed with caution given the large gaps in the evidence base. Of particular importance is the true incidence of typical symptomatic pertussis. If it is substantially higher than estimated in the base case analyses, as suggested by recent studies,31 herd immunity, and indeed most other factors, are less relevant. In that case, addition of an adolescent booster dose dominates current immunization practices. If incidence is not as high, then the extent of herd immunity becomes critical, and there are limited data with which to accurately predict this indirect benefit of an adolescent immunization strategy. Epidemiologic models attempting to simulate this herd effect have been made,17,31 but given available data, these efforts remain largely speculative. Although adolescent immunization programs have recently been implemented in France and Germany, pertussis is not a reportable disease in either country. Enhanced surveillance in countries such as these provides the best opportunity to fill the data gaps, but it will be many years before these data become available, if ever. Thus, the methods used to estimate inputs in the current evaluation will remain necessary, despite their imprecision.
Other strongly influential sources of uncertainty are the true incidence of pertussis, as well as the health and cost consequences of pertussis cases that go unreported. For example, if one accepts that <1% of cases are reported30 and that a fair proportion of these cases have meaningful clinical and economic consequences, then almost any strategy to vaccinate against pertussis will be cost-effective, regardless of whether vaccination leads to herd immunity. A 1% reporting rate coupled with one-half of the unreported cases being clinically significant leads to savings from both the health care payer and societal perspectives, even if the herd immunity impact of vaccinating adolescents was only 2%. Estimates of underreporting in the literature vary greatly, and there is none regarding their potential health and cost consequences. The estimate used in the base case analysis (fewer than 8 unreported cases per reported case) is conservative compared with other estimates found in the literature.28–31
The model used for these analyses is a relatively simple cohort simulation of the occurrence of pertussis and its management. More sophisticated models aimed at elucidating the succession of immune and infectious states over time, according to naturally acquired and vaccine-induced immunity, have been created.17,38 Those “SEIR” (susceptible, exposed, infective, recovered) models attempt to deduce the unknown variables by solving differential equations calibrated to the known epidemiology of pertussis in a particular setting. They typically require extensive data and demand considerable resources to create and analyze. That approach was not followed here because it was not deemed justified given the data gaps and because a simpler, easier-to-understand model was expected to yield the same level of information.
A health economic assessment of adolescent immunization based on a complex mathematical model that estimated the range of possible herd immunity outcomes if an adolescent immunization strategy was to be implemented has been published.17 The predicted reduction in pertussis cases in groups younger than the adolescent cohort ranged from 0 to 100%; that is, the model was not able to rule out any possible herd immunity outcomes. This being the case, the resulting health economic results were highly uncertain, but in line with those reported here for the United States.
Two other economic analyses of adolescent immunization have also recently been published, both adopting simpler modeling approaches similar to those presented in this paper.11,13 A cost-benefit analysis11 undertaken in the United States evaluated 7 independent strategies for administering a pertussis DTaP vaccine, 1 of which was administering a booster to adolescents (10–19 years of age). Based on an estimated incidence of pertussis in adolescents of more than 400 per 100,000 person-years, the analysis suggested that this strategy would prevent 0.7–1.8 million pertussis cases during 10 years and, excluding the cost of vaccination, would lead to savings of US $0.6–$1.6 billion. A break-even cost for vaccination of almost $37 was estimated. The analysis did not factor in herd immunity but used a considerably higher estimate of incidence than in our analyses.
A Canadian economic model, designed to evaluate replacement of an existing program of dT vaccination of adolescents with one using a DTaP vaccine, used an underreporting factor of 9.13 Although herd immunity was not explicitly accounted for, the model did assume a secondary attack rate of 12%. That evaluation estimated that adding a DTaP vaccine would lead to incremental health care costs of $188 per discounted pertussis case avoided but would lead to savings from the societal perspective of $0.60 per adolescent per year.
All studies to date are hampered by a lack of reliable data. Better data will allow more accurate estimates to be obtained and should be collected as part of any new program of immunization. As these data become available, the cost effectiveness of adolescent immunization should be reevaluated to confirm that the strategy is economically efficient. The availability of better data will allow for greater precision using more sophisticated modeling approaches and may permit estimation of shifts in the age distribution of cases resulting from immunization; the possible effects of immunization on changing the dynamics of natural infection, boosting immunity in the population; and the extent of herd immunity effects.
In the meantime, our analyses show that even with conservative estimates of the incidence of typical pertussis, expanding pertussis immunization to include adolescents can be cost-effective provided reasonable levels of herd immunity are attained. If incidence is much higher, then the direct benefits would provide sufficient economic justification. Either way, adolescent vaccination has the potential to significantly reduce the burden of disease in both adolescent and in infants, where it is so devastating.
1. Guris D, Strebel PM, Bardenheier B, et al. Changing epidemiology of pertussis in the United States: increasing reported incidence among adolescents and adults, 1990–1996. Clin Infect Dis
2. Wirsing von König CH, Halperin S, Rittelmann M, Guiso N. Pertussis of adults and infants. Lancet Infect Dis
3. Baron S, Njamkepo E, Grimprel E, et al. Epidemiology of pertussis in French hospitals in 1993 and 1994: thirty years after routine use of vaccination. Pediatr Infect Dis J
4. Deen JL, Mink CM, Cherry JD, et al. Household contact study of Bordetella pertussis
infections. Clin Infect Dis
5. Wirsing von König CH, Postels-Multani S, Bock HL, Schmitt HJ. Pertussis in adults: frequency of transmission after household exposure. Lancet
6. Lee LH, Pichichero ME. Costs of illness due to Bordetella pertussis
in families. Arch Fam Med
7. Corbel MJ, Xing DKL. Current status of acellular pertussis vaccines. J Med Microbiol
8. Koplan JP, Schoenbaum SC, Weinstein MC, Fraser DW. Pertussis vaccine: an analysis of benefits, risks, and costs. N Engl J Med
9. Hinman AR, Koplan JP. Pertussis and pertussis vaccine: reanalysis of benefits, risks, and costs. JAMA
10. Ekwueme DU, Strebel PM, Hadler SC, Meltzer MI, Allen JW, Livengood JR. Economic evaluation of use of diptheria, tetanus, and acellular pertussis vaccine or diphtheria, tetanus, and whole-cell pertussis vaccine in the United States. Arch Pediatr Adolesc Med
11. Purdy KW, Hay JW, Botteman MF, Ward JI. Evaluation of strategies for use of acellular pertussis vaccine in adolescents and adults: a cost-benefit analysis. Clin Infect Dis
12. Iskedjian M, Einarson TR, O'Brien BJ, et al. Economic evaluation of a new acellular vaccine for pertussis in Canada. Pharmacoeconomics
13. Iskedjian M, Walter JH, Hemels ME. Economic evaluation of an extended acellular pertussis vaccination programme for adolescents in Ontario, Canada. Vaccine
14. Tormans G, Van Doorslaer E, Van Damme P, Clara R, Schmitt HJ. Economic evaluation of pertussis prevention by whole-cell and acellular vaccine in Germany. Eur J Pediatr
15. Beutels P, Bonanni P, Tormans G, Canale F, Crovari PC. Economic evaluation of universal pertussis vaccination in Italy. Vaccine
16. Hudeckova H, Straka S, Szilagyiova M, Rusnakova S. Health and economic benefits of mandatory regular vaccination in the Slovak Republic, III: diphtheria, tetanus, and pertussis [in Slovak]. Epidemiol Mikrobiol Imunol
17. Edmunds WJ, Brisson M, Melegaro A, Gay NJ. Potential cost-effectiveness of acellular pertussis booster vaccination in England and Wales. Vaccine
18. Norlund A, Kornfalt R, Lithman T. Vaccination against whooping cough is beneficial for health economics: results of a model analysis [in Swedish]. Lakartidningen
19. Scuffham PA, McIntyre PB. Pertussis vaccination strategies for neonates: an exploratory effectiveness analysis. Vaccine
20. Weinstein MC, Siegel JE, Gold MR, Kamlet MS, Russell LB. Recommendations of the Panel of Cost-Effectiveness in Health and Medicine. JAMA
21. Centers for Disease Control and Prevention. Summary of notifiable diseases, United States, 1996. MMWR
22. Centers for Disease Control and Prevention. Summary of notifiable diseases, United States, 1997. MMWR
23. Centers for Disease Control and Prevention. Summary of notifiable diseases, United States, 1998. MMWR
24. Centers for Disease Control and Prevention. Summary of notifiable diseases, United States, 1999. MMWR
25. Black S. Epidemiology of pertussis. Pediatr Infect Dis J
26. Deeks S, De Serres G, Boulianne N, et al. Failure of physicians to consider the diagnosis of pertussis in children. Clin Infect Dis
27. Deville JG, Cherry JD, Christenson PD, et al. Frequency of unrecognized Bordetella pertussis
infections in adults. Clin Infect Dis
28. Miller E, Fleming DM, Ashworth LAE, Mabbett DA, Vurdien JE, Elliott TSJ. Serological evidence of pertussis in patients presenting with cough in general practice in Birmingham. Commun Dis Public Health
29. Nennig ME, Shinefield HR, Edwards KM, Black SB, Fireman BH. Prevalence and incidence of adult pertussis in an urban population. JAMA
30. Strebel P, Nordin J, Edwards K, et al. Population-based incidence of pertussis among adolescents and adults, Minnesota, 1995–1996. J Infect Dis
31. Van Rie A, Hethecote HW. Adolescent and adult pertussis vaccination: computer simulations of five new strategies. Vaccine
32. Sutter RW, Cochi SL. Pertussis hospitalizations and mortality in the United States, 1985–1989: evaluation of the completeness of national reporting. JAMA
33. Centers for Disease Control and Prevention. Pertussis: United States, 1997–2000. MMWR
34. Van der Wielen M, Van Damme P, Joossens E, Francois G, Meurice F, Ramalho A. Randomised controlled trial with a diphtheria-tetanus-acellular pertussis (dTpa) vaccine in adults. Vaccine
35. Ward J. Pertussis epidemiology and acellular pertussis vaccine efficacy in older children: NIH APERT multicenter pertussis trial. In: Pediatric Academic Societies' Meeting, Baltimore, MD, April 28–May 1, 2001
. Baltimore, MD: Pediatric Academic Societies; 2001. Abstract 1369.
36. Juretzko P, von Kries R, Hermann M, Wirsing von König CH, Giani G. Effectiveness of acellular pertussis vaccine assessed by hospital-based active surveillance in Germany. Clin Infect Dis
37. Salmaso S, Mastrantonio P, Tozzi AE, et al. Sustained efficacy during the first 6 years of life of 3-component acellular pertussis vaccines administered in infancy: the Italian experience. Pediatrics
38. Hethcote HW. Simulations of pertussis epidemiology in the United States: effects of adult booster vaccinations. Math Biosci
39. U.S. Department of Health and Human Services. Medicare and Medicaid statistical supplement, 2000: health care finance review, No. 03424. Baltimore, MD: HCFA Office of Strategic Planning; 2001.
40. St. Anthony's ICD-9-CM Code Book
. Reston, VA: St. Anthony Publishing, Inc.; 1998.
41. Office of Statewide Health Planning and Development. California 1999 Discharge Data
. version A.
42. State of Florida, Agency for Health Care Administration, State Center for Health Statistics. Florida 1999 Hospital Patient Data File
43. Maryland 1999 Inpatient Public Use File
. Glen Bernie, MD: St. Paul Computer Center, Inc.
44. Massachusetts Division of Health Care Finance and Policy. Massachusetts Fiscal Year 1999 Acute Hospital Case Mix Database
45. Washington State Department of Health, Office of Hospital and Patient Data. Washington 1999 CHARS (Comprehensive Hospital Abstract Reporting System) Public Data File
46. Bureau of Emergency Medical Services. Utah Emergency Department Encounter Database, 1999 Public-Use Data File
. Salt Lake City, UT: Utah Department of Health.
47. Minnesota Department of Health. Minnesota Case Mix Research Database, 2000
: Quality Assurance and Review Program
48. American Medical Association. Physicians' Current Procedural Terminology
. Chicago, IL: American Medical Association; 2002.
49. 2002 Clinical Diagnostic Laboratory Fee Schedule (CLAB) Public Use File (PUF)
. Centers for Medicare and Medicaid Services; 2002.
51. Verrilli DK, Zuckerman S. Preferred provider organizations and physician fees. Health Care Financ Rev
52. Medical Economics Staff. Medical Economics 2002 Drug Topics Red Book: The Pharmacist's Trusted Companion for More Than a Century
. Montvale, NJ: Medical Economics; 2002.
53. Accessed from the U.S. Bureau of Labor Statistics website. Available at: http://www.bls.gov
. Accessed June 1, 2002.
54. Denniston PL Jr, ed. Official Disability Guidelines, 1999
. 4th ed. Corpus Christi, TX: Work Loss Data Institute; 1998.
55. Denniston PL Jr, ed. Official Disability Guidelines, 2001
. 6th ed. Corpus Christi, TX: Work Loss Data Institute; 2000.
56. Denniston PL Jr, Ranavaya MI, eds. Official Disability Guidelines, 2002
. 7th ed. Corpus Christi, TX: Work Loss Data Institute; 2001.
57. De Serres G, Shadmani R, Duval B, et al. Morbidity of pertussis in adolescents and adults. J Infect Dis