Roberts, Scott MD, MS; Hollier, Lisa M. MD, MPH; Sheffield, Jeanne MD; Laibl, Vanessa MD; Wendel, George D. Jr MD
Epidemics of influenza are responsible for up to 36,000 deaths in the United States each year. Rates of infection are highest among children, but rates of serious illness and death are highest among persons aged 65 years or older and persons of any age who have medical conditions that place them at increased risk for complications from influenza. The risks of severe illness from influenza infection are elevated among both young children and pregnant women, and both groups benefit from vaccination by preventing illness and death from influenza. Influenza-associated excess deaths among pregnant women were documented during the pandemics of 1918–19 and 1957–58.1–4 Pregnancy can increase the risk for serious medical complications of influenza.5–9 An increased risk might result from 1) increases in heart rate, stroke volume, and oxygen consumption; 2) decreases in lung capacity; and 3) changes in immunologic function during pregnancy.
Since 2004 the Centers for Disease Control and Prevention (CDC) has recommended universal vaccination for all pregnant women, regardless of gestational age.10 There are no reports of perinatal morbidity from the use of inactivated trivalent vaccine, at least 2 reports of documented safety, and 1 report that implies potential harm from fever-induced teratogenicity in the unimmunized gravida.10–14 Differing conclusions have been reached concerning the cost-effectiveness of administering influenza vaccine to the general population.15,16 However, the average rate of hospital admission for influenza-related illness in these general working populations during the flu season was generally low: 20–40/100,000.17,18 The same is not true of the pregnant woman, where influenza-related hospitalizations run as high as 500–900/100,000.19,20 We sought to estimate whether universal influenza vaccination in pregnancy was cost-effective in the management of influenza-like illness during influenza season.
MATERIALS AND METHODS
All pregnant women aged 18–44 years in the United States were entered into a decision analysis model as a hypothetical cohort for 1 year. We adhered to the reference case guidelines of the Panel on Cost-Effectiveness in Health and Medicine.21 The analysis was conducted from the societal perspective and includes all relevant costs except secondary transmission of the influenza virus. In our effectiveness equations, we assumed that no deaths would occur, biasing the analysis in favor of the supportive care option.
Influenza-like illness is defined here as subjectively determined fever or a measured temperature of more than 37.7°C plus cough or sore throat (the World Health Organization definition).22 Although some studies with disparate definitions of influenza-like illness were used in this analysis, we adjusted the data by using different plausible rates and costs as they appeared in studies used.15
Influenza-like illness is a constellation of upper and lower respiratory conditions; therefore, it is associated with a high rate of illness. Although 40–50% of the civilian, noninstitutionalized population in the United States contracts a disease that meets this symptomatic definition during the course of an influenza season, the incidence rate of influenza virus infections is much lower, averaging less than 10 cases per 100 persons per year.15,23–27 Similar rates of influenza infection are reported in pregnant women.9,20
Supportive care only includes symptomatic treatment such as fluids, analgesics, and antitussives. No treatment specific for the flu such as oseltamivir or amantadine before hospitalization (if necessary) is included. Costs for medications received during hospitalization are included in hospitalization costs.
A decision analysis model was constructed using TreeAge Pro 2004 8.1 (TreeAge Software, Inc., Williamstown, MA). We examined 2 strategies: 1) vaccination of all pregnant women with the inactivated trivalent vaccine for 1 influenza season and 2) provision of supportive care only. The basic structure of the model was outlined by Muennig.28 The model was designed to obtain the probabilistically weighted average cost and effectiveness of each strategy by use of the inputs listed in Table 1. Each strategy was associated with a similar pathway of events: 1) the probability of illness, 2) the probability of a medical visit (including the probability of illness secondary to a primary influenza virus infection), 3) the probability of receiving antibiotics for secondary illnesses, 4) the probability of receiving over-the-counter medications, and 5) the probability of hospitalization. In the vaccination arm the probability of adverse effects was included, as were the costs of each treatment. A decision tree in simple format is included in Figure 1.
The costs associated with hospitalization caused by influenza virus infection and its complications were obtained from the 2002 Healthcare Cost and Utilization Project, a weighted database that includes information about approximately one half of all hospital discharges in the United States.41 Charges were inflated to 2004 charges by using the consumer price index for medical care (303.6/279.6, 2004/2002). Because hospital charges may not reflect the true societal costs of hospitalization, charges were converted to costs by use of cost-to-charge ratios derived from the Medical Provider Analysis and Review System found at The Centers for Medicare and Medicaid Services.29 Wages lost from work and leisure time in the hospital and transportation savings were included ($3853 + $1278 − $3.27) in hospitalization charges. The cost of an ambulatory care visit was determined by inflating the physician costs ($42.34 + $10.00 copay) as determined by Bridges et al15 using the consumer price index for medical care (2004/1999). The cost of routine influenza vaccination is assumed to be the cost of medication administration at a routine prenatal visit: The base-case cost for the vaccine and its administration used for the present analysis was $10.00. It was assumed that this visit would not cost extra because it was a part of routine prenatal care. We estimated antibiotic and over-the-counter drug consumption and costs by use of weighted mean values from randomized controlled trials published in the medical literature.15,30
Transportation use was estimated on the basis of data from the Bureau of the Census.31 Microcosting was effected using travel to work for the number of persons driving alone, carpooling, using public transportation, walking, and using other means ($1.72). Patient time included time spent in transit to a medical clinic or hospital in addition to the time spent receiving ambulatory care (1.5 hours at $17.75/hour) or hospital services. The cost of an ambulatory visit for influenza-like illness was: $1.72 + $42.34 + $10.00 + $26.63 = $80.69. Caregiver costs ($49.70) were obtained form the medical literature and the Bureau of Labor Statistics.42,32 All costs were adjusted to 2004 U.S. dollars.
The mean rate of hospitalization was obtained from the medical literature and local data concerning maternal influenza (Sheffield J, Laibl V, Roberts S, Lorimer M. The efficacy and safety of influenza vaccination in pregnancy [abstract]. Am J Obstet Gynecol 2004;191:S66).6,20 Data concerning pregnant women and influenza are limited but encompass both U.S. and Canadian cohorts. From the medical literature we obtained the following: 1) medical visits in vaccinated and unvaccinated persons,6,15,16 2) adverse effects from vaccination,10 3) Guillain-Barré syndrome33 resulting from influenza vaccination, and 4) secondary infection arising as a result of influenza.
The efficacy of influenza vaccine against influenza-like illness was calculated as a weighted mean from studies in healthy workers because data was lacking for pregnant women.15,16 A value of 0.70 was used as the efficacy of trivalent vaccine against influenza as reported by the CDC.10
We calculated quality-adjusted life years under the assumption that no deaths would occur among pregnant women. In fact, no deaths were reported in the Healthcare Cost and Utilization Project database for several years in the studied age group of women. The Quality of Well Being scale was used to estimate the health-related quality of life for persons with influenza-like illness.34 The Quality of Well Being scale converts data pertaining to mobility, physical activity, and social dimensions of a disease, along with information about the symptoms and problems a disease produces, to an health-related quality of life score. The Years of Healthy Life measure was used to assess the baseline health status of persons in the United States.43
Baseline measurement estimates and the range of plausible values for each estimate are listed in Table 1. All variables were tested for their influence on the model by a multivariate sensitivity (“tornado”) analysis. Variables that demonstrated sufficient influence on the rank order of the cost or effectiveness of each intervention were then tested in 1-way and, where appropriate, bivariate sensitivity analyses.
We also conducted a Monte Carlo simulation under the assumption that all variables would be triangular in distribution. The triangular distribution is a probability distribution in which the baseline value of a measure is assigned the highest probability of occurrence and the extreme high and low values of a measure are assigned the lowest probability of occurrence. The probability of observing a value between the baseline value and either the high or low value assigned to a variable is linearly interpolated. The Monte Carlo simulation was performed for 1 influenza season. The primary outcomes, total cost, total effectiveness, incremental cost, incremental effectiveness, and cost-effectiveness, are presented in standard format in Table 2.
Table 2 lists the average cost per person and effectiveness of the 2 interventions under study. The decision analysis model predicted that the strategy of vaccinating pregnant women aged 18–44 years was dominant (Figure 2). As can be seen in this figure, the vaccination strategy is more cost-effective for every probability of influenza-like illness. It costs more per quality of life year (quality-adjusted life year) with supportive care only. Vaccinating 50% of pregnant women aged 18–44 years (probability of influenza-like illness = 0.64) would result in a net gain of 0.00256 quality-adjusted life years or 22.2 quality-adjusted hours and savings of approximately $25 per pregnant woman relative to providing supportive care only during the flu season (Table 2). Vaccinating all pregnant women aged 18–44 years old would result in a net gain of 0.00512 quality-adjusted life years or 44.5 quality-adjusted hours and savings of approximately $50 per pregnant women relative to providing supportive care only.
When tested for costs, the model was most sensitive to the following variables, in decreasing order of sensitivity: incidence of influenza-like illness, incidence of influenza-like illness after influenza vaccination, probability of vaccination, and vaccine efficacy against influenza-like illness. Vaccination remained the dominant strategy when each of these variables was tested over the range of plausible values of each measure (Table 1).
The cost-effectiveness of the vaccination strategy was robust with respect to changes in the plausible values of all variables. None of the assumptions made in the analysis affected the dominance of the vaccination strategy when tested over a broad range of values. The Monte Carlo analysis predicted that vaccination would be a cost-saving strategy in 100% of all trials. Bridges et al15 reported a 50% rate of vaccine efficacy against laboratory-confirmed influenza in a healthy working population with a vaccine poorly matched to circulating virus strain in the 1997–1998 season. Even when vaccine is poorly matched to current influenza strains, cost savings are still realized (probability of vaccination = 0.5, vaccine efficacy against influenza = 0.5, vaccine efficacy against influenza-like illness = 0.28, cost savings of $22.69).
The costs of vaccination in this analysis only included the costs of the influenza vaccine given at the time of a routine prenatal visit. If all patients were to visit their obstetric care givers separately to receive the influenza vaccine during influenza season, vaccination would be associated with an incremental cost-effectiveness ratio of $7,563 per quality-adjusted life year relative to supportive care only. The cost-effectiveness ratio tells us how much it costs to save 1 year of healthy life relative to other interventions aimed at treating or preventing the same disease. Commonly accepted benchmarks for cost-effectiveness ratios are less than $50,000 per quality-adjusted life year. One recent cost-effectiveness analysis determined that the incremental cost-effectiveness ratio of vaccinating (with inactivated trivalent vaccine) all healthy adults aged 15–65 years old was $31,081 relative to supportive care.44
Vaccinating pregnant women aged 18–44 years with the inactivated trivalent influenza vaccine during influenza season is cost-saving relative to supportive care only. Vaccination of all persons in this age group would save approximately $50 and gain 0.00512 quality-adjusted life years per person relative to supportive care. Initially, we believe a cost savings associated with approximately 50% compliance is achievable based upon recent data suggesting that 44% of obstetricians offered pregnant women influenza vaccination when the CDC recommended routine vaccination in the second and third trimester of pregnancy.35
Our study is limited by a number of factors. Besides the usual uncertainties associated with cost-effectiveness analysis, we face a more serious uncertainty in cost and probability estimation because of the paucity of data concerning influenza in the pregnant woman. Many, if not most, of the measures used here are obtained from healthy working adults. One obvious difference between healthy working adults and the pregnant woman is that hospitalization rates are much higher for influenza during the influenza season in the pregnant population. Tuiyishime et al20 reported a hospitalization rate of 900 per 100,000 in a Canadian pregnant cohort and Hartert et al19 reported a hospitalization rate of 500 per 100,000 in the third trimester in similar pregnant women in a Tennessee Medicaid cohort. These rates are in sharp contrast to those in healthy working adults where rates of admission seldom exceed 20 to 40 per 100,000 persons unless accompanied by high-risk conditions such as asthma and aged older than 65 years.10 The higher reported rates were used in the sensitivity analyses, and the cost-effectiveness of the vaccination strategy was robust to variation in the rate of hospitalization.
Only Tuyishime et al20 has described the incidence of influenza-like illness in a pregnant cohort and found it to be 0.64. This is higher than reported in nonpregnant populations and contributes to most of the variation in our model, as it does in others.44 Regardless of the incidence of influenza-like illness, however, vaccination of the pregnant woman remained cost saving. This is most certainly due to the low cost of vaccination if administered during routine prenatal care.
We used the Quality of Well Being scale to derive the health-related quality of life for persons with influenza-like illness. This instrument is based on category scaling measurements. We chose this instrument because the health dimensions it incorporates are similar to those produced by influenza-like illness. Health-related quality of life scores typically assume a value from 0 to 1.0, with 0 being equivalent to death and 1.0 equivalent to perfect health. The Quality of Well Being scale was designed to assign a value to a particular disease, and it cannot be used to estimate the overall health of a population. For this reason, we used the Years of Healthy Life measure to estimate the overall health of our cohort for persons who did not develop an influenza-like illness and for persons who recover from an influenza-like illness. Morbidity during influenza-like illness is assigned into the health-related quality of life score. To the extent that morbid costs are incurred during the illness, not including hospitalization, these are reflected in the health-related quality of life score rather than costs in the numerator of the overall measure of cost-effectiveness. Lost wages and transportation savings during the illness for influenza-like illness during home convalescence would be reflected in quality-adjusted life years rather than cost savings.
This analysis was undertaken to reflect the most recent recommendations of the CDC. An unfortunate truth is that only 44% of practicing obstetricians follow the recommendations as they applied to influenza vaccination.35 Cost saving in a low risk (eg, patients without chronic respiratory, cardiac, or diabetic disorders) pregnant population implies cost-effectiveness in the higher-risk population. That is true in this model, where a prevalence of hospital admission of only 0.003 (in low-risk pregnant patients) as opposed to 0.005 (including those with chronic respiratory conditions) does not change the cost-savings associated with universal vaccination. A more rigorous analysis including previous recommendations of the CDC that called for universal immunization in the second and third trimesters of pregnancy and immunization in those pregnant women with chronic medical conditions45 or only immunization at any time in those with chronic medical conditions might demonstrate similar cost-savings or effectiveness.46 The authors are not convinced that superior cost-effectiveness or savings will be demonstrated. We hope the presentation of these results, a simpler model, will appeal to more obstetricians and increase the use of immunization guidelines in pregnancy as they currently apply.
Decreased illness in young infants is a potential yet currently theoretic and unquantifiable benefit to maternal influenza immunization.12,47–49 To the extent that it exists, the design presented remains societal but biased. Only an underappreciation of cost savings or effectiveness is possible by its exclusion.
In 2004, the CDC changed its recommendations from routine vaccination in the second and third trimester of pregnancy to universal vaccination during any trimester during pregnancy in the influenza season.10 We hope this shift will imply a sense of safety concerning the vaccine both to physician and pregnant patient alike. There are no confirmed fetal risks from use of the inactivated trivalent vaccine in pregnancy.10–13 This change should also provide impetus to obstetric providers to realize the importance of influenza vaccination. If pregnant women received influenza vaccination during influenza seasons, both cost-savings and additional quality-adjusted life would be realized.
1. Noble GR. Epidemiological and clinical aspects of influenza. In: Beare AS, editor. Basic and applied influenza research. Boca Raton (FL): CRC Press; 1982. p. 11-50.
2. Harris JW. Influenza occurring in pregnant women: a statistical study of thirteen hundred and fifty cases. JAMA 1919;72:978–80.
3. Widelock D, Csizmas L, Klein S. Influenza, pregnancy, and fetal outcome. Public Health Rep 1963;78:1–11.
4. Freeman DW, Barno A. Deaths from Asian influenza associated with pregnancy. Am J Obstet Gynecol 1959;78:1172–5.
5. Shahab SZ, Glezen WP. Influenza virus. In: Gonik B, editor. Viral diseases in pregnancy. New York (NY): Springer-Verlag; 1994. p. 215–23.
6. Schoenbaum SC, Weinstein L. Respiratory infection in pregnancy. Clin Obstet Gynecol 1979;22:293–300.
7. Kirshon B, Faro S, Zurawin RK, Samo TC, Carpenter RJ. Favorable outcome after treatment with amantadine and ribavirin in a pregnancy complicated by influenza pneumonia: a case report. J Reprod Med 1988;33:399–401.
8. Kort BA, Cefalo RC, Baker VV. Fatal influenza A pneumonia in pregnancy. Am J Perinatol 1986;3:179–82.
9. Irving WL, James DK, Stephenson T, Laing P, Jameson C, Oxford JS, et al. Influenza virus infection in the second and third trimesters of pregnancy: a clinical and seroepidemiological study. BJOG 2000;107:1282–9.
10. Harper SA, Fukuda K, Uyeki TM, Cox NJ, Bridges CB; Advisory Committee on Immunization Practices (ACIP), Centers for Disease Control and Prevention (CDC). Prevention and Control of Influenza. Recommendations of the Advisory Committee on Immunization Practices (ACIP) [Published erratum appears in MMWR Morb Mortal Wkly Rep. 2005;54:750]. MMWR Recomm Rep 2005;54(RR-8):1–40.
11. Immunization during pregnancy. ACOG Committee Opinion No. 282. American College Of Obstetricians and Gynecologists. Obstet Gynecol 2003;101:207–12.
12. Munoz FM, Greisinger AJ, Wehmanen OA, Mouzoon ME, Hoyle JC, Smith FA, et al. Safety of influenza vaccination during pregnancy. Am J Obstet Gynecol 2005;192:1098–106.
13. Heinonen OP, Shapiro S, Monson RR, Hartz SC, Rosenberg L, Slone D. Immunization during pregnancy against poliomyelitis and influenza in relation to childhood malignancy. Int J Epidemiol 1973;2:229–35.
14. Acs N, Banhidy F, Puho E, Czeizel AE. Maternal influenza during pregnancy and risk of congenital abnormalities in offspring. Birth Defects Res A Clin Mol Teratol 2005;73:989–96.
15. Bridges CB, Thompson WW, Meltzer MI, Reeve GR, Talamonti WJ, Cox NJ, et al. Effectiveness and cost-benefit of influenza vaccination of healthy working adults: a randomized controlled trial. JAMA 2000;284:1655–63.
16. Nichol KL, Lind A, Margolis KL, Murdoch M, McFadden R, Hauge M, et al. The effectiveness of vaccination against influenza in healthy, working adults. N Engl J Med 1995;333:889–93.
17. Simonsen L, Fukuda K, Schonberger LB, Cox NJ. Impact of influenza epidemics on hospitalizations. J Infect Dis 2000;181:831–7.
18. Barker WH, Mullooly JP. Impact of epidemic type A influenza in a defined adult population. Am J Epidemiol 1980;112:798–811.
19. Hartert TV, Neuzil KM, Shintani AK, Mitchel EF Jr, Snowden MS, Wood LB, et al. Maternal morbidity and perinatal outcomes among pregnant women with respiratory hospitalizations during influenza season. Am J Obstet Gynecol 2003;189:1705–12.
20. Tuyishime JD, De Wals P, Moutquin JM, Frost E. Influenza-like illness during pregnancy: results from a study in the eastern townships, Province of Quebec. J Obstet Gynaecol Can 2003;25:1020–5.
21. Gold MR, Siegel JE, Russel LB, Weinstein MC, editors. Cost-effectiveness in health and medicine. New York (NY): Oxford University Press; 1996.
22. Kendal AP, Pereria MS, Skehel JL. Concepts and procedures for laboratory-based influenza surveillance. Geneva (Switzerland): World Health Organization; 1982.
23. Nichol KL. Clinical effectiveness and cost effectiveness of influenza vaccination among healthy working adults. Vaccine 1999;17:S67–73.
24. Hayden FG, Osterhaus A, Treanor JJ, Fleming DM, Aoki FY, Nicholson KG, et al. Efficacy and safety of the neuraminidase inhibitor zanamivir in the treatment of influenza virus infections. GG167 Influenza Study Group. N Engl J Med 1997;337:874–80.
25. Monto AS, Robinson DP, Herlocher ML, Hinson JM Jr, Elliott MJ, Crisp A. Zanamivir in the prevention of influenza among healthy adults: a randomized controlled trial. JAMA 1999;282:31–5.
26. Wilde JA, McMillan JA, Serwint J, Butta J, O'Riordan MA, Steinhoff MC. Effectiveness of influenza vaccine in health care professionals: a randomized trial. JAMA 1999;281:908–13.
27. Harlan WR, Murt HA, Thomas W, Lepkowski JM, et al. Incidence, utilization, and costs associated with acute respiratory conditions. In: National medical care utilization and expenditure survey. Hyattsville (MD): U.S. Public Health Service, National Center for Health Statistics, Division of Health Interview Statistics; 1986:4–7.
28. Muennig P. Designing and conducting cost-effectiveness analyses in medicine and health care. San Francisco (CA): Jossey-Bass; 2002. p. 167–89.
30. Treanor JJ, Hayden FG, Vrooman PS, Barbarash R, Bettis R, Riff D, et al. Efficacy and safety of the oral neuraminidase inhibitor oseltamivir in treating acute influenza: a randomized controlled trial. US Oral Neuraminidase Study Group. JAMA 2000;283:1016–24.
32. Keech M, Scott AJ, Ryan PJ. The impact of influenza and influenza-like illness on productivity and healthcare resource utilization in a working population. Occup Med (Lond) 1998;48:85–90.
33. Lasky T, Terracciano GJ, Magder L, Koski CL, Ballesteros M, Nash D, et al. The Guillain-Barré syndrome and the 1992-1993 and 1993-1994 influenza vaccines. N Engl J Med 1998;339:1797–802.
34. Kaplan RM, Anderson JP. A general health policy model: update and applications. Health Serv Res 1988;23:203–35.
35. Schrag SJ, Fiore AE, Gonik B, Malik T, Reef S, Singleton JA, et al. Vaccination and perinatal infection prevention practices among obstetrician-gynecologists. Obstet Gynecol 2003;101:704–10.
36. Mauskopf JA, Cates SC, Griffin AD. A pharmacoeconomics model for the treatment of influenza. Pharmacoeconomics 1999;16:73–84.
37. Nichol KL. Cost-benefit analysis of a strategy to vaccinate healthy working adults against influenza. Arch Intern Med 2001;161:749–59.
38. Monto AS, Ohmit SE, Margulies JR, Talsma A. Medical practice-based influenza surveillance: viral prevalence and assessment of morbidity. Am J Epidemiol 1995;141:502–6.
39. Adams PF, Hendershot GE, Marano MA; Centers for Disease Control and Prevention/National Center for Health Statistics. Current estimates from the national health interview survey, 1996. Vital Health Stat 10 1999;10:1–212.
40. Campbell DS, Rumley MH. Cost-effectiveness of the influenza vaccine in a healthy, working-age population. J Occup Environ Med 1997;39:408–14.
42. U.S. Department of Labor. Bureau of Labor Statistics. National Compensation Survey. NCS: Occupational Wages in the United States, Bulletin, 2002. Available at: http://www.bls.gov/ncs/ocs/sp/ncb10552.pdf
. Retrieved October 15, 2005.
43. Gold MR, Franks P, McCoy KI, Fryback DG. Toward consistency in cost-utility analyses: using national measures to create condition-specific values. Med Care 1998;36:778–92.
44. Muennig PA, Khan K. Cost-effectiveness of vaccination versus treatment of influenza in healthy adolescents and adults. Clin Infect Dis 2001;33:1879–85.
45. Centers for Disease Control and Prevention. Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep 1997;46(RR-9): 1–5.
46. Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep 1993;42(RR-6):5.
47. Neuzil KM, Mellen BG, Wright PF, Mitchel EF Jr, Griffin MR. The effect of influenza on hospitalizations, outpatient visits, and courses of antibiotics in children. N Engl J Med 2000;342:225-31.
48. Neuzil KM, Reed GW, Mitchel EF, Simonsen L, Griffin MR. Impact of influenza on acute cardiopulmonary hospitalizations in pregnant women. Am J Epidemiol 1998;148:1094–102.
49. Puck JM, Glezen WP, Frank AL, Six HR. Protection of infants from infection with influenza A virus by transplacentally acquired antibody. J Infect Dis 1980;142:844–9.