- Email: [email protected]
Cost-effectiveness of the introduction of the pneumococcal polysaccharide vaccine in elderly Colombian population
Vaccine 29 (2011) 7644–7650
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Cost-effectiveness of the introduction of the pneumococcal polysaccharide vaccine in elderly Colombian population a,∗ ˜ Carlos Castaneda-Orjuela , Nelson Alvis-Guzmán b,1 , Ángel José Paternina a,b,1 , Fernando De la Hoz-Restrepo a a Epidemiology and Public Health Evaluation Group, Department of Public Health, Faculty of Medicine, Universidad Nacional de Colombia, Carrera 30 # 45-03, Office 150, Bogotá D.C., Colombia b Department of Economic and Social Research-DIES, Universidad de Cartagena, Colombia, Calle 70 No. 7-33, Cartagena de Indias, Colombia
a r t i c l e
i n f o
Article history: Received 25 March 2011 Received in revised form 27 July 2011 Accepted 1 August 2011 Available online 30 August 2011 Keywords: Streptococcus pneumoniae Vaccine Cost–benefit analysis Aged Colombia
a b s t r a c t Background: Streptococcus pneumoniae causes community-acquired pneumonia, otitis media and meningitis, with higher incidences at the extremes of life. PPV-23 vaccine is widely used in prevention of pneumonia and invasive pneumococcal disease in older adults in developed countries. We developed an evaluation of cost-effectiveness of implementing PPV-23 in Colombian population over 60 years. Methods: The number of cases of pneumonia and meningitis in patients over 60 years and the proportion by S. pneumoniae was estimated based on a review of literature. A decision tree model with a 5-year time horizon was built to evaluate the cost-effectiveness of the implementation of the PPV-23 in this population. Direct health care costs of out- and in-patients were calculated based on expenditure records from the Bogota public health system. Incremental cost-effectiveness ratios per life saved and per year of life gained were estimated based on the decision tree model. Deterministic and probabilistic sensitivity analyses were performed. Results: Without vaccination 4460 (range 2384–8162) bacteremic pneumococcal pneumonias and 141 (range 73–183) pneumococcal meningitis would occur among people over 60 years old in Colombia. In the first year, vaccination with PPV-23 at US$8/dose would save 480 (range 100–1753) deaths due to Invasive and non-invasive pneumococcal disease. Vaccination would results in US$3400/deaths averted (range US$1028–10,862) and US$1514/life years gained (range US$408–5404). Conclusion: Vaccination with PPV-23 in over 60 years is a highly cost-effective public health measure in Colombia. Despite some limitations, the results are robust, and may help developing countries to perform informed decisions about the introduction of the vaccine. © 2011 Elsevier Ltd. All rights reserved.
1. Introduction Streptococcus pneumoniae is an important pathogen that causes community-acquired pneumonia (CAP), otitis media and meningitis that affects children and adults worldwide [1,2]. It is associated in 30–50% of all CAP, with the highest incidence in children under 2 years and adults over 65 years [3–6]. The World Health Organization (WHO) estimated that in 2002 around 1.6 million deaths
Abbreviations: CAP, community-acquired pneumonia; IPD, invasive pneumococcal disease; MoH, Ministry of Health; PPV-23, pneumococcal polysaccharide vaccine of 23 serotypes; WHO, World Health Organization. ∗ Corresponding author. Tel.: +57 3165000x15086; fax: +57 3165175. E-mail addresses: [email protected], [email protected] ˜ [email protected] (N. Alvis-Guzmán), (C. Castaneda-Orjuela), [email protected] (Á.J. Paternina), [email protected] (F. De la Hoz-Restrepo). 1 Tel.: +57 3157434337. 0264-410X/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.vaccine.2011.08.006
occurred worldwide associated to pneumococcal disease, mostly in children and the elderly [7]. Developed countries like the United States, Canada, Finland, Sweden, Denmark, Norway, England and Israel, reported an annual incidence rate in older adults (over 65 years) from 24 to 85 cases of invasive pneumococcal disease (IPD) per 100,000 people [1]. However, in developing countries, like Colombia, there are no population-based studies that estimate the incidence of syndromes related S. pneumoniae, which hampers the estimation of the burden of disease and lead to delays in the process of deciding which public health measures should be undertaken to prevent the disease. There is indirect evidence that infection by S. pneumoniae later in life may be an important problem in Colombia. Syndromes partially related to pneumococcal were an important cause of mortality among Colombian eldest in 2007. Pneumonia, meningitis and sepsis from all causes accounted for 4.4% of all deaths occurring in people over 60 years old at that year. Conversely, deaths due to pneumonia among people above 60 represented 66.5% of deaths by
C. Casta˜ neda-Orjuela et al. / Vaccine 29 (2011) 7644–7650
pneumonia across all age groups. For meningitis and sepsis, deaths among the elderly represented 28.4% and 58.0% of all deaths by such causes respectively [8]. Given the scarcity of epidemiological information researchers need to combine multiple data sources (health care records, databases of mortality and surveys of services access) to estimate the epidemiological impact of pneumococcal disease. Currently, there are four vaccines against S. pneumoniae available. One is a polysaccharide vaccine of 23 serotypes (PPV-23) while the other three are conjugate vaccines that contain 7, 10 and 13 serotypes. The polysaccharide vaccine has been used extensively in developed countries showing an intermediate level of protection against invasive pneumococcal disease in people over 65 years [9,10]. Also, it has been recommended for very high-risk persons, including persons with severe spleen dysfunction, and immunocompromised patients [9,11,12]. PPV-23 is less used in developing countries partly due to the scarcity of epidemiological evidence for its introduction. Systematic reviews have shown that in developing countries, PPV-23 vaccine used in older adults is highly effective for reducing mortality due to pneumonia, while in developed countries the effect is not always clear [13–17]. These studies emphasized in differences in methodology, populations used in the different analysis, and highlighted the limitation of randomized control trials in determine the effectiveness of vaccination. In general, studies supporting the effectiveness against invasive pneumococcal disease, especially in elderly high-risk patients, are based on evidence of observational designs. This article shows the results of a cost-effectiveness analysis of the use of PPV-23 in older adults of a middle-income South American country, Colombia, where the vaccine is not currently offered by the public health system. The study was commissioned by the Colombian Ministry of Health (MoH) in order to estimate the potential benefits of the PPV-23 vaccine intervention to prevent deaths in people over 60 years old. The results of the study would help the MoH and other health decision makers to reach an informed decision over the PPV-23 introduction.
2. Materials and methods 2.1. Structure of the decision model For the cost-effectiveness analysis a decision tree model was built (Fig. 1), based on the one used by De Graeve et al. [18], which simulated two alternatives: (a) use of one dose of PPV-23 in adults over 60 years or (b) no use of PPV-23. The model did not consider potential adverse effects of PPV-23. Transition from one state to other was determined by vaccine coverage, vaccine efficacy and risk of pneumococcal infection in people over 60 years. Epidemiological inputs of the model were extracted from a previous exhaustive search for information, and a review of health care records within the Colombian Health System that have been published elsewhere [19]. The model was based on the following assumptions: (1) vaccinated people who were not protected by the vaccine had the same chance of being infected than those not vaccinated. (2) The maximum protection for the vaccine occurred after 15 days of the first dose. (3) The probability of occurrence of the categories of pneumonia and meningitis are allocated based on previous estimates of the proportions of cases due to pneumococcal. (4) Only pneumonia and meningitis were included as clinical outcomes of S. pneumoniae infection. (5) The final states are healing or death. The time horizon was 5 years and both, cost estimates and outcomes were annually discounted at a rate of 3%. Life years gained (LYG) was used as the main outcome measures.
7645
Fig. 1. Decision tree model to evaluate the cost-effectiveness of PPV-23 vaccine in aged. Adapted from De Graeve et al. [18].
2.2. Epidemiological parameters The population over 60 years, at risk of developing pneumococcal diseases in Colombia, was taken from projections of the National Census carried out by the National Bureau of Statistics (DANE) in 2008 [20]. The mortality rate for older adults caused by respiratory diseases was estimated from the mortality database of Bogota DC administered by the Bogota’s Health Services. Bogota DC is the capital of Colombia and concentrates more than 15% of the country population. They were selected, because they are better in coverage, completeness and accuracy than the national mortality records managed by DANE. Additionally a validation of this data was carrying out to extrapolate this information to national level based on the MoH registers and adjustments of the national mortality reports. The crossing of the various databases allowed us to compare and validate the estimates with data from Bogota. The frequency of pneumonia and meningitis caused by S. pneumoniae was estimated using several steps. First, the rate of general consultation (outpatient attention by any cause) for people over 60 years was estimated using two sources: (a) the database of the Individual Records of Delivery of Health Care (Registro individual de Prestación de Servicios, RIPS), and (b) the Colombian National Demographic and Health Survey, 2005 (CNDHS) [21]. In a second step, the proportion of attention due to pneumonia and meningitis by all agents was estimated by calculating the proportion of Health Care System discharges by both causes. For pneumonia ICD 10 codes from J10 to J18 were taken in account while for meningitis ICD 10 codes G000–G039 were used. This proportion was estimated from a sample of RIPS records and then was multiplied by the rate of consultation by all cause calculated in the first step. Third, the result of the previous step for pneumonia was multiplied by the proportion of pneumococcal pneumonia found in the literature from population-based studies. Thus, the yearly risk of pneumococcal pneumonia was estimated and a similar approach was used for meningitis. Then, those risks were incorporated in the decision tree model (see Figure Supplementary 1). These parameters were compared and validated with results from studies found in the search of data in Latin American countries. We searched in Pubmed, Scielo and LILACS databases with the following MeSH terms: “Pneumonia”, “Meningitis”, “Sepsis”, “Streptococcus pneumoniae”, “pneumococcus”, “Cost of Illness”, “burden of disease”, “Prevalence”, “Incidence”, “Hospitalization”, “mortality”, “Mortality”, “epidemiology”, (“Americas” NOT “North
7646
C. Casta˜ neda-Orjuela et al. / Vaccine 29 (2011) 7644–7650
Table 1 Parameters used in the model. Parameter Epidemiological parameters Annual incidence of all-agents pneumonias by 100.000 % Fatality of all-agents pneumonias Annual mortality rate by 100,000 of all-agents pneumonias % of pneumonias due to S. pneumoniae Annual incidence by 100,000 of all-agents meningitis % of meningitis due to S. pneumoniae % Fatality of all-agents meningitis Attention costs Outpatient pneumonia attention Inpatient pneumonia attention Inpatient meningitis attention Vaccination Population > 60 Efficacy against all causes pneumonia Efficacy against pneumococcal pneumonia Efficacy against pneumococcal meningitis Vaccination coverage Costs of PPV-23 schedule (US$2008)
Mean estimate 1023.1 29.0 218 .8 10 .5 11.4 29.0% 54.7% $61 $1020 $1221 4,151,533 10% 50% 52% 70% 8
America”) OR “Mexico”), (“Age Groups” NOT “Infants”), in different combinations and with its Spanish equivalents. We consider only population-based studies, with occurrence original data, blood or normally sterile liquid culture as diagnostic criteria, and that provide data in older adults populations. Studies with sample less than 100 subjects and published before 1990 were excluded. We identified 7 studies [22–28] of which we used their estimations as parameters of pneumococcal disease occurrence (pneumonia and meningitis) in older than 60 years. When more than one estimate was reported or when there were two studies for the same estimator was calculated median and interquartile ranges. Input parameters are shown in Table 1. The coverage vaccination assumption was taken based on Colombian experience in vaccination of elderly population with influenza vaccine. A discount rate of 3% to both costs and outcomes was applied. 2.3. Vaccine’s effectiveness The value for parameters of efficacy and effectiveness of the PPV-23 vaccine used in the economic model were those reported by a meta-analysis conducted by Moberley et al. [17]. This study reported an effectiveness of PPV of 74% (CI 95%: 54–85%) against IPD in adults from RCT, with better effect in low-income countries (86%, IC95%: 59–97%). Against all-cause pneumonia the effectiveness was 29% (IC95%: 3–48%) in adults and 46% (IC95%: 33–57%) in low-income countries. When were considered the observational studies the PPV effectiveness against IPD was 68% (IC95%: 53–78%) in immunocompetent older adults. Based in these estimations we assumed a PPV-23 effectiveness of 50% (40–60%) against invasive pneumococcal pneumonia, 52% (37–61%) against (invasive) pneumococcal meningitis and 10% (5–15%) against all-cause pneumonias (Table 1). This scenario is compatible with usual PPV-23 effectiveness assumption in older people [29]. The parameter effectiveness against all-cause pneumonia allowed estimating cases of non-invasive pneumococcal pneumonia averted with vaccination to cross with invasive pneumonias data. 2.4. Costs parameters Monetary values for cost of care (in- and out-patient direct medical cost for pneumonia and inpatient costs for meningitis) were estimated from a sample of clinical records (ambulatory and
Inferior limit
Superior limit
Distributional parameters
Reference
797.6 19 .6 88.4 7.2 8.0 19.6% 51.4%
1435.0 38.4 536.3 13.7 12.8 38.4% 59.5%
Beta (1.6, 3) Beta (3, 3) Not applied Beta (3.1, 3) Beta (7.3, 3) Beta (4.4, 3) Beta (2.1, 3)
[19] [23] [24] [22–23] [25–26] [27–28] [26]
$60 $560 $1215
$204 $2222 $2185
Beta (2, 6) Beta (1.1, 3) Beta (2, 6)
RIPS-MPS RIPS-MPS RIPS-MPS
4,151,533 5% 40% 37% 50% 5
4,151,533 15% 60% 61% 90% 10
Not allowed Beta (3, 3) Beta (3, 3) Beta (5, 3) Beta (3,3) Beta (4.5, 3)
DANE [46] [46] [46] MPS MPS
hospitalary) of the RIPS database (Bogota’s Health Service) with a diagnosis of pneumonia or meningitis (ICD-10 J10–J18 for pneumonia and G000–G039 for meningitis). Besides clinical data on diagnosis, RIPS records also have data on the cost of every intervention that have been administered to patients allowing researchers to estimate the direct medical cost of disease, without the cost of outpatient pharmacological treatment. The perspective of the analysis was from third payer (Colombian Health System). The costs of treating each case of invasive pneumococcal disease, either pneumonia or meningitis, are shown in Table 1. Given that RIPS did not discriminate diagnosis of pneumonia or meningitis by specific agents, it had to be assumed that the direct medical costs of those syndromes associated to S. pneumoniae were similar to those caused by other agents. For pneumonia, the estimated direct medical costs ranged between US$60 and US$203 (point estimate US$61.4) per ambulatory managed patient and US$1020.5 (range US$560.5–2221.6) for hospital managed case. Meningitis was assumed to be treated only in the hospital setting and the estimated cost per patient ranged from US$1215–2185 (point estimate US$1220). The proportion of hospitalization found in the review of the case database to estimate costs was 55% for pneumonia and 100% for meningitis. The cost of PPV-23 dose was taking of MoH communication that corresponds to purchase and administration costs. All costs are expressed in US 2008 dollars. 2.5. Cost-effectiveness estimations To analyse the cost-effectiveness was used incremental costeffectiveness ratio (ICER) defined as: ICER =
Costs associated with vaccination − treatment costs avoided Life years gained
2.6. Sensitivity analysis To consider the uncertainty of the epidemiological and economical parameters used for the model a deterministic univariate and probabilistic multivariate sensitivity analysis were conducted. The variables included in this step were: the cost of the vaccine, vaccination coverage in the studied population, and the effectiveness of the vaccine to prevent cases and deaths. Acceptability curves of vaccination and no vaccination alternatives were constructed.
C. Casta˜ neda-Orjuela et al. / Vaccine 29 (2011) 7644–7650
7647
Table 2 Burden of pneumococcal disease in over 60 years old, Colombia, during the time horizon analysis. Parameter Without vaccination All-causes pneumonias Invasive pneumococcal pneumonias Pneumococcal meningitis All-cause pneumonia deaths Invasive pneumococcal pneumonia deaths Pneumococcal meningitis deaths Total deaths Years of life lost (without discount) Years of life lost (with discount) With vaccination All-causes pneumonias Invasive pneumococcal pneumonias Pneumococcal meningitis All-cause pneumonia deaths Invasive pneumococcal pneumonia deaths Pneumococcal meningitis deaths Total deaths Years of life lost (without discount) Years of life lost (with discount)
Mean estimate
Inferior limit
Superior limit
211,107 22,166 701 61,222 6428
165,045 11,883 380 32,347 2328
294,592 40,360 905 113,122 15,498
384 61,606 154,382 138,070
195 32,542 81,457 65,435
539 113,661 285,414 285,414
196,411 21,910 445 56,959 4181
160,929 11,813 310 31,541 1863
255,203 39,863 445 97,997 7137
245 57,204 143,327 128,183
160 31,701 79,353 63,744
245 97,240 246,544 246,544
3. Results In 2008, the Colombian population over 60 years was 4,151,533 people (DANE). In the absence of PPV-23 vaccination it was estimated that in the first year 42,474 cases would occur of all cause pneumonia (range 33,113–59,574), and 4460 (range 2384–8162) of them would be associated to a bacteremic infection by S. pneumoniae. For pneumococcal meningitis the expected number of cases amounted to 141 (range 76–183). The predicted number of deaths by all cause pneumonia was expected to be between 6490 and 22,876 with 1293 (range 467–3134) associated to S. pneumoniae bacteremic infection. Pneumococcal meningitis would account for 77 deaths (range 39–109). The accumulated results for the 5 years of analysis are shown in Table 2. The overall direct medical costs associated to invasive diseases by S. pneumoniae during the period of analysis was estimated at US$12,839,220 (range US$3,772,500–55,401,556) with discount (Table 3) During the time horizon of the analysis, the cases avoided after the introduction of the PPV-23 vaccine, at a price per dose of US$8, would amount 14,696 (range 4116–39,389) invasive and non-invasive pneumonias. For meningitis it was calculated that 256 (range 70–497) would be avoided after vaccination. Thus, the costs of invasive pneumococcal disease burden in the vaccination scenario are US$8,336,070 (range US$3,024,098–25,501,613) with discount. When we consider not only invasive disease, but also non-invasive pneumonias the total costs of treating those diseases was US$115,599,627 (range US$47,382,762–391,922,083) without vaccination and US$107,317,612 (range US$46,139,069–338,701,649) when vaccinating. The total avoided costs by decreasing the burden of disease amount to US$8,282,016 (range US$1,243,693–53,220,434) with discount, or US$9,038,842 (range US$1,475,721–53,220,434) without discount setting. The deaths averted amounted to 4402 (range 841–15,421), of which 2247 (range 465–8361) were saved to IPD. The total LYG amounted to 11,055 (range 2105–38,870) over the time horizon (5 years), and 9887 (range 1691–38,870) after the discounting. The incremental cost to prevent one death with the vaccination strategy is US$3,400 (range US$1028–10,862), and gaining
Fig. 2. Deterministic sensitivity analysis ICER US$/LYG.
a life year would cost US$1514 (range US$408–5404) (Table 3). Without discounting the ICER per LYG will be US$1285 (range US$408–4231). The deterministic univariate sensitivity analysis (Fig. 2) shows that the cost per LYG in the 5-year time horizon is more sensitive to changes in effectiveness against all cause pneumonia, vaccine cost, incidence of all-cause pneumonia, care costs of pneumonia inpatient and the probability of death from all causes pneumonia. Acceptability curves (Fig. 3) show that at a willingness to pay over US$1500 per LYG is more likely that vaccination alternative been cost-effectiveness. 4. Discussion This is the first study that assesses the introduction of the PPV23 vaccine in a developing country. The results suggest that the use of the PPV-23 vaccine in the Colombian population over 60 years old is a highly cost-effective measure compared to the nonvaccination scenario. PPV-23 introduction would be justified by a willingness to pay above US$1,514/LYG over a time horizon of 5 years. Though Colombia has not set a cost-effectiveness threshold for adopting new health care strategies, WHO recommends a threshold at US$4799 (Colombia per capita Gross Domestic Product). Thus, the ICER of the intervention would be below the level recommended by WHO for countries where no specific value of willingness to pay has been estimated [30]. The PPV-23 vaccination in first 5 years avoids 6943 non-invasive pneumococcal pneumonias, 7753 invasive pneumococcal pneumonias and 256 pneumococcal meningitis and prevents 4402 deaths the first 5 years after vaccination. When we only considered the IPD in the analysis the ICER of vaccination versus status quo was US$3487/LYG (range US$327–9587), lower than the willingness to pay assumed in the analysis. The results are robust to significant parameters changes. With respect to serotype coverage of PPV-23, the Colombian National Health Institute (Instituto Nacional de Salud, INS) perform the surveillance of circulating serotypes of S. pneumoniae since 1994. The Network Surveillance System for the Bacterial Agents Responsible for Pneumonia and Meningitis (SIREVA II) strategy in Colombia in years 2006 to 2008 reported that in aged the 13 more frequent pneumococcal serotypes (1, 3, 4, 5, 6A, 6B, 7F, 9V, 14, 18C, 19A, 19F and 23F), account to 63.0% of invasive pneumococcal disease.
7648
C. Casta˜ neda-Orjuela et al. / Vaccine 29 (2011) 7644–7650
Table 3 Costs of burden of pneumococcal disease in over 60 years, Colombia, during the time horizon. Results with discount. Parameter
Mean estimate
Inferior limit
Superior limit
Without vaccination Costs of pneumonias (included pneumococcal invasive) Costs of invasive pneumococcal pneumonias Costs of pneumococcal meningitis Cost IPD (invasive pneumococcal pneumonias + pneumococcal meningitis) Total costs (all pneumonias + pneumococcal meningitis)
$114,815,745 $12,055,338 $783,882 $12,839,220 $115,599,627
$46,993,688 $3,383,426 $389,074 $3,772,500 $47,382,762
$389,944,375 $53,423,847 $1,977,709 $55,401,556 $391,922,083
With vaccination Vaccination costs Costs of pneumonias (included pneumococcal invasive) Costs of invasive pneumococcal pneumonias Costs of pneumococcal meningitis Total costs (vaccination plus burden of disease) Costs averted ICER per pneumonia case averted ICER per meningitis case averted ICER death averted ICER LYG without discount ICER LYG with discount
$23,248,584 $106,820,102 $7,838,560 $497,509 $130,566,196 $8,282,016 $1.018 $58.463 $3400 $1285 $1514
$10,378,835 $45,821,666 $2,706,695 $317,403 $56,517,904 $1,243,693 −$403 −$31.905 −$1028 −$408 −$408
$37,363,800 $337,810,041 $24,610,005 $891,608 $376,065,449 $53,220,434 $2.219 $130.502 $10,862 $4231 $5404
In developed countries, most studies have found that PPV-23 vaccination is a cost-effective measure and even a cost-saving in people older than 65 years old [29,31–33]. In a review by Ogilvie et al. [29] of analyses in 16 developed countries, Sweden was the only country with the US$ per quality adjusted life year above the willingness to pay (US$50,000). However, those studies show that uncertainty about several epidemiologic and economic parameters may affect the ICER. The most important variables affecting cost-effectiveness are the costs of vaccination, efficacy of the vaccine, IPD incidence and case-fatality rate [29]. In France in 1981, Levy found that PPV-23 was costeffective when applied to a population over 45 years [34]. Patrick and Woolley in 1981 showed the cost-benefits of the vaccine in people over 50 years and those with high-risk conditions (chronically cardiovascular, kidney, or liver ill, and diabetes), recommending the application in this population after to compare 3 alternatives: no vaccination, vaccinate all or vaccinate only high risk [35]. They considered a vaccine cost of US$8.61 per dose with polysaccharide vaccine of 14 serotypes plus operative costs, and only evaluate the pneumococcal pneumonia outcome. In 1997, the Advisory Committee on Immunization Practices (ACIP) updated the guidelines for recommending pneumococcal vaccine to vaccinate the entire population over 65 years, immunocompromised patients and high-risk populations [9].
In 2002, a cost-effectiveness study for people over 65 years in Catalonia (Spain) showed that pneumococcal vaccination for the population over 65 years was a cost-saving strategy to a cost per dose of D 11.51 [36]. Authors noted, that pneumococcal vaccination for older people was as cost-effective as other preventive interventions in this population, such as hypercholesterolemia, hypertension treatment, and smoking cessation therapies [36]. This study has limitations. There are no Colombian data for some of the epidemiological parameters related to the frequency of pneumococcal disease in adults above 60 years. Therefore, those indicators had to be estimated from the Latin American medical literature, which could potentially limit some of the conclusions. Despite that, the proportion of all pneumonias that could be associated to pneumococcal infection assumed in this analysis (10.5%) is a conservative estimate that may lead to a potential underestimation of the burden of disease and the benefits of pneumococcal vaccination [37–39]. Similarly population based studies from developed countries have found rates of all-cause pneumonia higher than 10% [40]. Also, important variables, such as the efficacy against all-cause pneumonia, incidence of IPD and case-fatality ratio are unavailable for Colombia. This is not a problem unique to our study. A review of the cost-effectiveness of pneumococcal vaccination strategies [29], highlighted the wide variation in assumptions for the key variables, even within population subgroups.
Fig. 3. Acceptability curve of cost-effectiveness analysis of PPV-23 vaccination in elderly Colombian population.
C. Casta˜ neda-Orjuela et al. / Vaccine 29 (2011) 7644–7650
Other limitations of our study are the non-inclusion of indirect effects of pneumococcal disease that underestimated the benefits of vaccination, and the non-consideration of antimicrobial resistance. Although the present study did not include otitis media as an outcome, the low frequency of this event in the elderly population may have a small effect in the results, underestimating the effect of vaccination. On the other hand bacteremia by pneumococcal also did not included which could lead to underestimate vaccination effects, but this should partially avoided when we included pneumonias and meningitis with blood isolation of S. pneumoniae. The efficacy of PPV-23 against all cause pneumonia was one of the most sensitive variables to changes. At efficacy values less than 4% PPV-23 is not cost-effective. The efficacy of PPV-23, for non-invasive pneumococcal pneumonia, has been always a highly controversial issue, especially in elderly population. The last WHO position paper on use of PPV-23 (2008) [41], recognizes that the evidence for efficacy and effectiveness in children and adults is poor, based on a recent review by Moberley and Huss in 2008 and 2009 respectively [17,42]. However it still acknowledges effectiveness from 50 to 80% for IPD prevention among adults [15]. In addition, the WHO paper highlights the existing challenges to estimate accurately the effectiveness of PPV-23 on the field, including the low frequency of IPD, the lack of consensus on diagnostic criteria for pneumococcal pneumonia, the variation of the effectiveness with age, and the severity of several underlying conditions associated with an increased risk of pneumococcal disease [41]. It is also recognized that very few randomized trials have been carried out and those who have done could have methodological problems that can potentially biased the results [15,43–45]. Despite the scarcity of evidence, WHO recommends PPV-23 vaccination in developed and developing countries given the high burden of pneumococcal disease among adults and children [41]. Despite of above descript limitations our analysis results are important as evidence for Colombian decision makers. However, other approach as budget impact analysis and political considerations can be important to evaluate in the Colombian and other developing countries settings. 5. Conclusion This analysis found that vaccination in the population over 60 year olds with the PPV-23 vaccine could be a very cost-effective strategy for the prevention of IPD in Colombia, a South American developing country. The results are consistent when the effect of non-invasive pneumococcal pneumonias is excluded. A surveillance analysis will be needed to evaluate the real impact of PPV-23 vaccination in the Colombian setting after its implementation. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.vaccine.2011.08.006. References [1] Fedson DS, Scott JA. The burden of pneumococcal disease among adults in developed and developing countries: what is and is not known. Vaccine 1999;17(July (Suppl. 1)):S11–8. [2] 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(April (13)):1729–35. [3] Ortqvist A, Hedlund J, Kalin M. Streptococcus pneumoniae: epidemiology, risk factors, and clinical features. Semin Respir Crit Care Med 2005;26(December (6)):563–74. [4] Monto AS. Acute respiratory infection in children of developing countries: challenge of the 1990s. Rev Infect Dis 1989;11(May–June (3)):498–505.
7649
[5] Burman LA, Norrby R, Trollfors B. Invasive pneumococcal infections: incidence, predisposing factors, and prognosis. Rev Infect Dis 1985;7(March–April (2)):133–42. [6] Marrie TJ, Durant H, Yates L. Community-acquired pneumonia requiring hospitalization: 5-year prospective study. Rev Infect Dis 1989;11(July–August (4)):586–99. [7] World Health Organization. Estimates of disease burden and costeffectiveness. Geneva; 2008 [cited 2010 junio]. Available from: http://www.who.int/immunization monitoring/burden/estimates burden/en/ index.html. [8] DANE. Cuadro 5. Defunciones, por grupos de edad y sexo, ˜ 2007. según lista de causas agrupadas 6/67 CIE-1 de OPS. Ano Bogota, D.C.: DANE; 2007 [cited 2011 May]. Available from: http://www.dane.gov.co/daneweb V09/index.php?option=com content&view =article&id=205&Itemid=119. [9] Prevention of pneumococcal disease: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep 1997;46(April (RR-8)):1–24. [10] Bogaert D, Hermans PW, Adrian PV, Rumke HC, de Groot R. Pneumococcal vaccines: an update on current strategies. Vaccine 2004;22(June (17–18)):2209–20. [11] NHS. In: NHS, editor. Pneumococcal vaccine for older people. London, UK: Department of Health; 2004. [12] Dhankhar P, Grabenstein JD, O’Brien MA, Dasbach EJ. Cost-effectiveness of stockpiling 23-valent pneumococcal polysaccharide vaccine to prevent secondary pneumococcal infections among a high-risk population in the United States during an influenza pandemic. Clin Ther 2010;32(August (8)):1501–16. [13] Watson L, Wilson BJ, Waugh N. Pneumococcal polysaccharide vaccine: a systematic review of clinical effectiveness in adults. Vaccine 2002;20(May (17–18)):2166–73. [14] Moore RA, Wiffen PJ, Lipsky BA. Are the pneumococcal polysaccharide vaccines effective? Meta-analysis of the prospective trials. BMC Fam Pract 2000;1:1. [15] Conaty S, Watson L, Dinnes J, Waugh N. The effectiveness of pneumococcal polysaccharide vaccines in adults: a systematic review of observational studies and comparison with results from randomised controlled trials. Vaccine 2004;22(August (23–24)):3214–24. [16] Melegaro A, Edmunds WJ. The 23-valent pneumococcal polysaccharide vaccine. Part I. Efficacy of PPV in the elderly: a comparison of meta-analyses. Eur J Epidemiol 2004;19(4):353–63. [17] Moberley SA, Holden J, Tatham DP, Andrews RM. Vaccines for preventing pneumococcal infection in adults. Cochrane Database Syst Rev 2008;1:CD000422. [18] De Graeve D, Lombaert G, Goossens H. Cost-effectiveness analysis of pneumococcal vaccination of adults and elderly persons in Belgium. Pharmacoeconomics 2000;17(June (6)):591–601. [19] Castaneda-Orjuela C, Alvis-Guzman N, de la Hoz-Restrepo F. The impact of pneumococcal disease on adults living in Bogota, Colombia, 2008. Rev Salud Publica (Bogota) 2010;12(February (1)):38–50. [20] DANE. Proyecciones de Población Municipales. Bogotá, D.C.: DANE; 2009 [2009 Feb 23]. Available from: http://www.dane.gov.co/index.php?option=com content&task=category§ionid=16&id=497&Itemid=995. [21] PROFAMILIA. Encuesta Nacional de Demografia y Salud Bogota D.C.; 2005. [22] Matute AJ, Brouwer WP, Hak E, Delgado E, Alonso E, Hoepelman IM. Aetiology and resistance patterns of community-acquired pneumonia in Leon, Nicaragua. Int J Antimicrob Agents 2006;28(November (5)):423–7. [23] Diaz A, Vera C, Santamarina J, Izarduy L, Bagnesco R, Grinspon S, et al. Neumonía adquirida en la comunidad en acncianos que requieren internación.Cuadro clínico y pronóstico. Medicina (Arg) 1999;59(6):731–8. [24] Szot J. Mortalidad por enfermedades respiratorias en Chile durante 1999. Rev Chil Enferm Respir 2003;19(1):8–14. [25] Dickinson FO, Pérez AE. Meningoencefalitis bacterianas en Cuba. Rev Cubana Hig Epidemiol 2001;39(2):86–94. [26] Dickinson Meneses F, Pérez Rodríguez A. Meningoencefalitis bacterianas en el adulto mayor cubano 1998–2000. Geriátrika 2002;18(6):32–40. [27] Gomes I, Lucena R, Melo A. Clinical and laboratory characteristics of pyogenic meningitis in adults. Arq Neuropsiquiatr 1997;55(September (3B)): 584–7. [28] Rossoni AM, Dalla Costa LM, Berto DB, Farah SS, Gelain M, Brandileone MC, et al. Acute bacterial meningitis caused by Streptococcus pneumoniae resistant to the antimicrobian agents and their serotypes. Arq Neuropsiquiatr 2008;66(September (3A)):509–15. [29] Ogilvie I, Khoury AE, Cui Y, Dasbach E, Grabenstein JD, Goetghebeur M. Cost-effectiveness of pneumococcal polysaccharide vaccination in adults: a systematic review of conclusions and assumptions. Vaccine 2009;27(August (36)):4891–904. [30] WHO Commission on Macroeconomics and Health. Macroeconomics and health: investing in health for economic development. Report of the Commission on Macroeconomics and Health. Geneva; 2001. [31] Willems JS. Cost effectiveness and cost benefit of vaccination against pneumococcal pneumonia: implications for clinical practice. Infect Control 1982;3(July–August (4)):299–302. [32] Willems JS, Sanders CR, Riddiough MA, Bell JC. Cost effectiveness of vaccination against pneumococcal pneumonia. N Engl J Med 1980;303(September (10)):553–9. [33] Ament A, Baltussen R, Duru G, Rigaud-Bully C, de Graeve D, Ortqvist A, et al. Cost-effectiveness of pneumococcal vaccination of older people: a study in 5 western European countries. Clin Infect Dis 2000;31(August (2)):444–50.
7650
C. Casta˜ neda-Orjuela et al. / Vaccine 29 (2011) 7644–7650
[34] Levy E. Cost effectiveness of pneumococcal vaccine (author’s transl). Rev Epidemiol Sante Publique 1981;29(2):133–53. [35] Patrick KM, Woolley FR. A cost–benefit analysis of immunization for pneumococcal pneumonia. JAMA 1981;245(February (5)):473–7. [36] Plans P. Cost-effectiveness of 23-valent antipneumococcical vaccination in Catalonia (Spain). Gac Sanit 2002;16(September–October (5)):392–400. [37] Ruiz-Gonzalez A, Falguera M, Nogues A, Rubio-Caballero M. Is Streptococcus pneumoniae the leading cause of pneumonia of unknown etiology? A microbiologic study of lung aspirates in consecutive patients with community-acquired pneumonia. Am J Med 1999;106(April (4)):385–90. [38] Mandell LA, Wunderink RG, Anzueto A, Bartlett JG, Campbell GD, Dean NC, et al. Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community-acquired pneumonia in adults. Clin Infect Dis 2007;44(March (Suppl. 2)):S27–72. [39] Brown PD, Lerner SA. Community-acquired pneumonia. Lancet 1998;352(October (9136)):1295–302. [40] Jackson ML, Neuzil KM, Thompson WW, Shay DK, Yu O, Hanson CA, et al. The burden of community-acquired pneumonia in seniors: results of a populationbased study. Clin Infect Dis 2004;39(December (11)):1642–50.
[41] World Health Organization, 23-valent pneumococcal polysaccharide vaccine. WHO position paper. Wkly Epidemiol Rec 2008;83(October (42)): 373–84. [42] Huss A, Pippa Scott P, Stuck AE, Trotter C, Egger M. Efficacy of pneumococcal vaccination in adults: a meta-analysis. Can Med Assoc J 2009;180(January (1)):48–58. [43] Butler JC, Breiman RF, Campbell JF, Lipman HB, Broome CV, Facklam RR. Pneumococcal polysaccharide vaccine efficacy. An evaluation of current recommendations. JAMA 1993;270(October (15)):1826– 31. [44] Mangtani P, Cutts F, Hall AJ. Efficacy of polysaccharide pneumococcal vaccine in adults in more developed countries: the state of the evidence. Lancet Infect Dis 2003;3(February (2)):71–8. [45] Zhogolev SD, Ogarkov PI, Mel’nichenko PI. The prophylaxis of nonhospital pneumonia using 23-valent pneumococcus vaccine in the military collectives. Voen Med Zh 2004;325(December (12)):35–43, 96. [46] Standaert B, Parez N, Tehard B, Colin X, Detournay B. Cost-effectiveness analysis of vaccination against rotavirus with RIX4414 in France. Appl Health Economics Health Policy 2008;6(4):199–216.
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Cost-effectiveness of the introduction of the pneumococcal polysaccharide vaccine in elderly Colombian population a,∗ ˜ Carlos Castaneda-Orjuela , Nelson Alvis-Guzmán b,1 , Ángel José Paternina a,b,1 , Fernando De la Hoz-Restrepo a a Epidemiology and Public Health Evaluation Group, Department of Public Health, Faculty of Medicine, Universidad Nacional de Colombia, Carrera 30 # 45-03, Office 150, Bogotá D.C., Colombia b Department of Economic and Social Research-DIES, Universidad de Cartagena, Colombia, Calle 70 No. 7-33, Cartagena de Indias, Colombia
a r t i c l e
i n f o
Article history: Received 25 March 2011 Received in revised form 27 July 2011 Accepted 1 August 2011 Available online 30 August 2011 Keywords: Streptococcus pneumoniae Vaccine Cost–benefit analysis Aged Colombia
a b s t r a c t Background: Streptococcus pneumoniae causes community-acquired pneumonia, otitis media and meningitis, with higher incidences at the extremes of life. PPV-23 vaccine is widely used in prevention of pneumonia and invasive pneumococcal disease in older adults in developed countries. We developed an evaluation of cost-effectiveness of implementing PPV-23 in Colombian population over 60 years. Methods: The number of cases of pneumonia and meningitis in patients over 60 years and the proportion by S. pneumoniae was estimated based on a review of literature. A decision tree model with a 5-year time horizon was built to evaluate the cost-effectiveness of the implementation of the PPV-23 in this population. Direct health care costs of out- and in-patients were calculated based on expenditure records from the Bogota public health system. Incremental cost-effectiveness ratios per life saved and per year of life gained were estimated based on the decision tree model. Deterministic and probabilistic sensitivity analyses were performed. Results: Without vaccination 4460 (range 2384–8162) bacteremic pneumococcal pneumonias and 141 (range 73–183) pneumococcal meningitis would occur among people over 60 years old in Colombia. In the first year, vaccination with PPV-23 at US$8/dose would save 480 (range 100–1753) deaths due to Invasive and non-invasive pneumococcal disease. Vaccination would results in US$3400/deaths averted (range US$1028–10,862) and US$1514/life years gained (range US$408–5404). Conclusion: Vaccination with PPV-23 in over 60 years is a highly cost-effective public health measure in Colombia. Despite some limitations, the results are robust, and may help developing countries to perform informed decisions about the introduction of the vaccine. © 2011 Elsevier Ltd. All rights reserved.
1. Introduction Streptococcus pneumoniae is an important pathogen that causes community-acquired pneumonia (CAP), otitis media and meningitis that affects children and adults worldwide [1,2]. It is associated in 30–50% of all CAP, with the highest incidence in children under 2 years and adults over 65 years [3–6]. The World Health Organization (WHO) estimated that in 2002 around 1.6 million deaths
Abbreviations: CAP, community-acquired pneumonia; IPD, invasive pneumococcal disease; MoH, Ministry of Health; PPV-23, pneumococcal polysaccharide vaccine of 23 serotypes; WHO, World Health Organization. ∗ Corresponding author. Tel.: +57 3165000x15086; fax: +57 3165175. E-mail addresses: [email protected], [email protected] ˜ [email protected] (N. Alvis-Guzmán), (C. Castaneda-Orjuela), [email protected] (Á.J. Paternina), [email protected] (F. De la Hoz-Restrepo). 1 Tel.: +57 3157434337. 0264-410X/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.vaccine.2011.08.006
occurred worldwide associated to pneumococcal disease, mostly in children and the elderly [7]. Developed countries like the United States, Canada, Finland, Sweden, Denmark, Norway, England and Israel, reported an annual incidence rate in older adults (over 65 years) from 24 to 85 cases of invasive pneumococcal disease (IPD) per 100,000 people [1]. However, in developing countries, like Colombia, there are no population-based studies that estimate the incidence of syndromes related S. pneumoniae, which hampers the estimation of the burden of disease and lead to delays in the process of deciding which public health measures should be undertaken to prevent the disease. There is indirect evidence that infection by S. pneumoniae later in life may be an important problem in Colombia. Syndromes partially related to pneumococcal were an important cause of mortality among Colombian eldest in 2007. Pneumonia, meningitis and sepsis from all causes accounted for 4.4% of all deaths occurring in people over 60 years old at that year. Conversely, deaths due to pneumonia among people above 60 represented 66.5% of deaths by
C. Casta˜ neda-Orjuela et al. / Vaccine 29 (2011) 7644–7650
pneumonia across all age groups. For meningitis and sepsis, deaths among the elderly represented 28.4% and 58.0% of all deaths by such causes respectively [8]. Given the scarcity of epidemiological information researchers need to combine multiple data sources (health care records, databases of mortality and surveys of services access) to estimate the epidemiological impact of pneumococcal disease. Currently, there are four vaccines against S. pneumoniae available. One is a polysaccharide vaccine of 23 serotypes (PPV-23) while the other three are conjugate vaccines that contain 7, 10 and 13 serotypes. The polysaccharide vaccine has been used extensively in developed countries showing an intermediate level of protection against invasive pneumococcal disease in people over 65 years [9,10]. Also, it has been recommended for very high-risk persons, including persons with severe spleen dysfunction, and immunocompromised patients [9,11,12]. PPV-23 is less used in developing countries partly due to the scarcity of epidemiological evidence for its introduction. Systematic reviews have shown that in developing countries, PPV-23 vaccine used in older adults is highly effective for reducing mortality due to pneumonia, while in developed countries the effect is not always clear [13–17]. These studies emphasized in differences in methodology, populations used in the different analysis, and highlighted the limitation of randomized control trials in determine the effectiveness of vaccination. In general, studies supporting the effectiveness against invasive pneumococcal disease, especially in elderly high-risk patients, are based on evidence of observational designs. This article shows the results of a cost-effectiveness analysis of the use of PPV-23 in older adults of a middle-income South American country, Colombia, where the vaccine is not currently offered by the public health system. The study was commissioned by the Colombian Ministry of Health (MoH) in order to estimate the potential benefits of the PPV-23 vaccine intervention to prevent deaths in people over 60 years old. The results of the study would help the MoH and other health decision makers to reach an informed decision over the PPV-23 introduction.
2. Materials and methods 2.1. Structure of the decision model For the cost-effectiveness analysis a decision tree model was built (Fig. 1), based on the one used by De Graeve et al. [18], which simulated two alternatives: (a) use of one dose of PPV-23 in adults over 60 years or (b) no use of PPV-23. The model did not consider potential adverse effects of PPV-23. Transition from one state to other was determined by vaccine coverage, vaccine efficacy and risk of pneumococcal infection in people over 60 years. Epidemiological inputs of the model were extracted from a previous exhaustive search for information, and a review of health care records within the Colombian Health System that have been published elsewhere [19]. The model was based on the following assumptions: (1) vaccinated people who were not protected by the vaccine had the same chance of being infected than those not vaccinated. (2) The maximum protection for the vaccine occurred after 15 days of the first dose. (3) The probability of occurrence of the categories of pneumonia and meningitis are allocated based on previous estimates of the proportions of cases due to pneumococcal. (4) Only pneumonia and meningitis were included as clinical outcomes of S. pneumoniae infection. (5) The final states are healing or death. The time horizon was 5 years and both, cost estimates and outcomes were annually discounted at a rate of 3%. Life years gained (LYG) was used as the main outcome measures.
7645
Fig. 1. Decision tree model to evaluate the cost-effectiveness of PPV-23 vaccine in aged. Adapted from De Graeve et al. [18].
2.2. Epidemiological parameters The population over 60 years, at risk of developing pneumococcal diseases in Colombia, was taken from projections of the National Census carried out by the National Bureau of Statistics (DANE) in 2008 [20]. The mortality rate for older adults caused by respiratory diseases was estimated from the mortality database of Bogota DC administered by the Bogota’s Health Services. Bogota DC is the capital of Colombia and concentrates more than 15% of the country population. They were selected, because they are better in coverage, completeness and accuracy than the national mortality records managed by DANE. Additionally a validation of this data was carrying out to extrapolate this information to national level based on the MoH registers and adjustments of the national mortality reports. The crossing of the various databases allowed us to compare and validate the estimates with data from Bogota. The frequency of pneumonia and meningitis caused by S. pneumoniae was estimated using several steps. First, the rate of general consultation (outpatient attention by any cause) for people over 60 years was estimated using two sources: (a) the database of the Individual Records of Delivery of Health Care (Registro individual de Prestación de Servicios, RIPS), and (b) the Colombian National Demographic and Health Survey, 2005 (CNDHS) [21]. In a second step, the proportion of attention due to pneumonia and meningitis by all agents was estimated by calculating the proportion of Health Care System discharges by both causes. For pneumonia ICD 10 codes from J10 to J18 were taken in account while for meningitis ICD 10 codes G000–G039 were used. This proportion was estimated from a sample of RIPS records and then was multiplied by the rate of consultation by all cause calculated in the first step. Third, the result of the previous step for pneumonia was multiplied by the proportion of pneumococcal pneumonia found in the literature from population-based studies. Thus, the yearly risk of pneumococcal pneumonia was estimated and a similar approach was used for meningitis. Then, those risks were incorporated in the decision tree model (see Figure Supplementary 1). These parameters were compared and validated with results from studies found in the search of data in Latin American countries. We searched in Pubmed, Scielo and LILACS databases with the following MeSH terms: “Pneumonia”, “Meningitis”, “Sepsis”, “Streptococcus pneumoniae”, “pneumococcus”, “Cost of Illness”, “burden of disease”, “Prevalence”, “Incidence”, “Hospitalization”, “mortality”, “Mortality”, “epidemiology”, (“Americas” NOT “North
7646
C. Casta˜ neda-Orjuela et al. / Vaccine 29 (2011) 7644–7650
Table 1 Parameters used in the model. Parameter Epidemiological parameters Annual incidence of all-agents pneumonias by 100.000 % Fatality of all-agents pneumonias Annual mortality rate by 100,000 of all-agents pneumonias % of pneumonias due to S. pneumoniae Annual incidence by 100,000 of all-agents meningitis % of meningitis due to S. pneumoniae % Fatality of all-agents meningitis Attention costs Outpatient pneumonia attention Inpatient pneumonia attention Inpatient meningitis attention Vaccination Population > 60 Efficacy against all causes pneumonia Efficacy against pneumococcal pneumonia Efficacy against pneumococcal meningitis Vaccination coverage Costs of PPV-23 schedule (US$2008)
Mean estimate 1023.1 29.0 218 .8 10 .5 11.4 29.0% 54.7% $61 $1020 $1221 4,151,533 10% 50% 52% 70% 8
America”) OR “Mexico”), (“Age Groups” NOT “Infants”), in different combinations and with its Spanish equivalents. We consider only population-based studies, with occurrence original data, blood or normally sterile liquid culture as diagnostic criteria, and that provide data in older adults populations. Studies with sample less than 100 subjects and published before 1990 were excluded. We identified 7 studies [22–28] of which we used their estimations as parameters of pneumococcal disease occurrence (pneumonia and meningitis) in older than 60 years. When more than one estimate was reported or when there were two studies for the same estimator was calculated median and interquartile ranges. Input parameters are shown in Table 1. The coverage vaccination assumption was taken based on Colombian experience in vaccination of elderly population with influenza vaccine. A discount rate of 3% to both costs and outcomes was applied. 2.3. Vaccine’s effectiveness The value for parameters of efficacy and effectiveness of the PPV-23 vaccine used in the economic model were those reported by a meta-analysis conducted by Moberley et al. [17]. This study reported an effectiveness of PPV of 74% (CI 95%: 54–85%) against IPD in adults from RCT, with better effect in low-income countries (86%, IC95%: 59–97%). Against all-cause pneumonia the effectiveness was 29% (IC95%: 3–48%) in adults and 46% (IC95%: 33–57%) in low-income countries. When were considered the observational studies the PPV effectiveness against IPD was 68% (IC95%: 53–78%) in immunocompetent older adults. Based in these estimations we assumed a PPV-23 effectiveness of 50% (40–60%) against invasive pneumococcal pneumonia, 52% (37–61%) against (invasive) pneumococcal meningitis and 10% (5–15%) against all-cause pneumonias (Table 1). This scenario is compatible with usual PPV-23 effectiveness assumption in older people [29]. The parameter effectiveness against all-cause pneumonia allowed estimating cases of non-invasive pneumococcal pneumonia averted with vaccination to cross with invasive pneumonias data. 2.4. Costs parameters Monetary values for cost of care (in- and out-patient direct medical cost for pneumonia and inpatient costs for meningitis) were estimated from a sample of clinical records (ambulatory and
Inferior limit
Superior limit
Distributional parameters
Reference
797.6 19 .6 88.4 7.2 8.0 19.6% 51.4%
1435.0 38.4 536.3 13.7 12.8 38.4% 59.5%
Beta (1.6, 3) Beta (3, 3) Not applied Beta (3.1, 3) Beta (7.3, 3) Beta (4.4, 3) Beta (2.1, 3)
[19] [23] [24] [22–23] [25–26] [27–28] [26]
$60 $560 $1215
$204 $2222 $2185
Beta (2, 6) Beta (1.1, 3) Beta (2, 6)
RIPS-MPS RIPS-MPS RIPS-MPS
4,151,533 5% 40% 37% 50% 5
4,151,533 15% 60% 61% 90% 10
Not allowed Beta (3, 3) Beta (3, 3) Beta (5, 3) Beta (3,3) Beta (4.5, 3)
DANE [46] [46] [46] MPS MPS
hospitalary) of the RIPS database (Bogota’s Health Service) with a diagnosis of pneumonia or meningitis (ICD-10 J10–J18 for pneumonia and G000–G039 for meningitis). Besides clinical data on diagnosis, RIPS records also have data on the cost of every intervention that have been administered to patients allowing researchers to estimate the direct medical cost of disease, without the cost of outpatient pharmacological treatment. The perspective of the analysis was from third payer (Colombian Health System). The costs of treating each case of invasive pneumococcal disease, either pneumonia or meningitis, are shown in Table 1. Given that RIPS did not discriminate diagnosis of pneumonia or meningitis by specific agents, it had to be assumed that the direct medical costs of those syndromes associated to S. pneumoniae were similar to those caused by other agents. For pneumonia, the estimated direct medical costs ranged between US$60 and US$203 (point estimate US$61.4) per ambulatory managed patient and US$1020.5 (range US$560.5–2221.6) for hospital managed case. Meningitis was assumed to be treated only in the hospital setting and the estimated cost per patient ranged from US$1215–2185 (point estimate US$1220). The proportion of hospitalization found in the review of the case database to estimate costs was 55% for pneumonia and 100% for meningitis. The cost of PPV-23 dose was taking of MoH communication that corresponds to purchase and administration costs. All costs are expressed in US 2008 dollars. 2.5. Cost-effectiveness estimations To analyse the cost-effectiveness was used incremental costeffectiveness ratio (ICER) defined as: ICER =
Costs associated with vaccination − treatment costs avoided Life years gained
2.6. Sensitivity analysis To consider the uncertainty of the epidemiological and economical parameters used for the model a deterministic univariate and probabilistic multivariate sensitivity analysis were conducted. The variables included in this step were: the cost of the vaccine, vaccination coverage in the studied population, and the effectiveness of the vaccine to prevent cases and deaths. Acceptability curves of vaccination and no vaccination alternatives were constructed.
C. Casta˜ neda-Orjuela et al. / Vaccine 29 (2011) 7644–7650
7647
Table 2 Burden of pneumococcal disease in over 60 years old, Colombia, during the time horizon analysis. Parameter Without vaccination All-causes pneumonias Invasive pneumococcal pneumonias Pneumococcal meningitis All-cause pneumonia deaths Invasive pneumococcal pneumonia deaths Pneumococcal meningitis deaths Total deaths Years of life lost (without discount) Years of life lost (with discount) With vaccination All-causes pneumonias Invasive pneumococcal pneumonias Pneumococcal meningitis All-cause pneumonia deaths Invasive pneumococcal pneumonia deaths Pneumococcal meningitis deaths Total deaths Years of life lost (without discount) Years of life lost (with discount)
Mean estimate
Inferior limit
Superior limit
211,107 22,166 701 61,222 6428
165,045 11,883 380 32,347 2328
294,592 40,360 905 113,122 15,498
384 61,606 154,382 138,070
195 32,542 81,457 65,435
539 113,661 285,414 285,414
196,411 21,910 445 56,959 4181
160,929 11,813 310 31,541 1863
255,203 39,863 445 97,997 7137
245 57,204 143,327 128,183
160 31,701 79,353 63,744
245 97,240 246,544 246,544
3. Results In 2008, the Colombian population over 60 years was 4,151,533 people (DANE). In the absence of PPV-23 vaccination it was estimated that in the first year 42,474 cases would occur of all cause pneumonia (range 33,113–59,574), and 4460 (range 2384–8162) of them would be associated to a bacteremic infection by S. pneumoniae. For pneumococcal meningitis the expected number of cases amounted to 141 (range 76–183). The predicted number of deaths by all cause pneumonia was expected to be between 6490 and 22,876 with 1293 (range 467–3134) associated to S. pneumoniae bacteremic infection. Pneumococcal meningitis would account for 77 deaths (range 39–109). The accumulated results for the 5 years of analysis are shown in Table 2. The overall direct medical costs associated to invasive diseases by S. pneumoniae during the period of analysis was estimated at US$12,839,220 (range US$3,772,500–55,401,556) with discount (Table 3) During the time horizon of the analysis, the cases avoided after the introduction of the PPV-23 vaccine, at a price per dose of US$8, would amount 14,696 (range 4116–39,389) invasive and non-invasive pneumonias. For meningitis it was calculated that 256 (range 70–497) would be avoided after vaccination. Thus, the costs of invasive pneumococcal disease burden in the vaccination scenario are US$8,336,070 (range US$3,024,098–25,501,613) with discount. When we consider not only invasive disease, but also non-invasive pneumonias the total costs of treating those diseases was US$115,599,627 (range US$47,382,762–391,922,083) without vaccination and US$107,317,612 (range US$46,139,069–338,701,649) when vaccinating. The total avoided costs by decreasing the burden of disease amount to US$8,282,016 (range US$1,243,693–53,220,434) with discount, or US$9,038,842 (range US$1,475,721–53,220,434) without discount setting. The deaths averted amounted to 4402 (range 841–15,421), of which 2247 (range 465–8361) were saved to IPD. The total LYG amounted to 11,055 (range 2105–38,870) over the time horizon (5 years), and 9887 (range 1691–38,870) after the discounting. The incremental cost to prevent one death with the vaccination strategy is US$3,400 (range US$1028–10,862), and gaining
Fig. 2. Deterministic sensitivity analysis ICER US$/LYG.
a life year would cost US$1514 (range US$408–5404) (Table 3). Without discounting the ICER per LYG will be US$1285 (range US$408–4231). The deterministic univariate sensitivity analysis (Fig. 2) shows that the cost per LYG in the 5-year time horizon is more sensitive to changes in effectiveness against all cause pneumonia, vaccine cost, incidence of all-cause pneumonia, care costs of pneumonia inpatient and the probability of death from all causes pneumonia. Acceptability curves (Fig. 3) show that at a willingness to pay over US$1500 per LYG is more likely that vaccination alternative been cost-effectiveness. 4. Discussion This is the first study that assesses the introduction of the PPV23 vaccine in a developing country. The results suggest that the use of the PPV-23 vaccine in the Colombian population over 60 years old is a highly cost-effective measure compared to the nonvaccination scenario. PPV-23 introduction would be justified by a willingness to pay above US$1,514/LYG over a time horizon of 5 years. Though Colombia has not set a cost-effectiveness threshold for adopting new health care strategies, WHO recommends a threshold at US$4799 (Colombia per capita Gross Domestic Product). Thus, the ICER of the intervention would be below the level recommended by WHO for countries where no specific value of willingness to pay has been estimated [30]. The PPV-23 vaccination in first 5 years avoids 6943 non-invasive pneumococcal pneumonias, 7753 invasive pneumococcal pneumonias and 256 pneumococcal meningitis and prevents 4402 deaths the first 5 years after vaccination. When we only considered the IPD in the analysis the ICER of vaccination versus status quo was US$3487/LYG (range US$327–9587), lower than the willingness to pay assumed in the analysis. The results are robust to significant parameters changes. With respect to serotype coverage of PPV-23, the Colombian National Health Institute (Instituto Nacional de Salud, INS) perform the surveillance of circulating serotypes of S. pneumoniae since 1994. The Network Surveillance System for the Bacterial Agents Responsible for Pneumonia and Meningitis (SIREVA II) strategy in Colombia in years 2006 to 2008 reported that in aged the 13 more frequent pneumococcal serotypes (1, 3, 4, 5, 6A, 6B, 7F, 9V, 14, 18C, 19A, 19F and 23F), account to 63.0% of invasive pneumococcal disease.
7648
C. Casta˜ neda-Orjuela et al. / Vaccine 29 (2011) 7644–7650
Table 3 Costs of burden of pneumococcal disease in over 60 years, Colombia, during the time horizon. Results with discount. Parameter
Mean estimate
Inferior limit
Superior limit
Without vaccination Costs of pneumonias (included pneumococcal invasive) Costs of invasive pneumococcal pneumonias Costs of pneumococcal meningitis Cost IPD (invasive pneumococcal pneumonias + pneumococcal meningitis) Total costs (all pneumonias + pneumococcal meningitis)
$114,815,745 $12,055,338 $783,882 $12,839,220 $115,599,627
$46,993,688 $3,383,426 $389,074 $3,772,500 $47,382,762
$389,944,375 $53,423,847 $1,977,709 $55,401,556 $391,922,083
With vaccination Vaccination costs Costs of pneumonias (included pneumococcal invasive) Costs of invasive pneumococcal pneumonias Costs of pneumococcal meningitis Total costs (vaccination plus burden of disease) Costs averted ICER per pneumonia case averted ICER per meningitis case averted ICER death averted ICER LYG without discount ICER LYG with discount
$23,248,584 $106,820,102 $7,838,560 $497,509 $130,566,196 $8,282,016 $1.018 $58.463 $3400 $1285 $1514
$10,378,835 $45,821,666 $2,706,695 $317,403 $56,517,904 $1,243,693 −$403 −$31.905 −$1028 −$408 −$408
$37,363,800 $337,810,041 $24,610,005 $891,608 $376,065,449 $53,220,434 $2.219 $130.502 $10,862 $4231 $5404
In developed countries, most studies have found that PPV-23 vaccination is a cost-effective measure and even a cost-saving in people older than 65 years old [29,31–33]. In a review by Ogilvie et al. [29] of analyses in 16 developed countries, Sweden was the only country with the US$ per quality adjusted life year above the willingness to pay (US$50,000). However, those studies show that uncertainty about several epidemiologic and economic parameters may affect the ICER. The most important variables affecting cost-effectiveness are the costs of vaccination, efficacy of the vaccine, IPD incidence and case-fatality rate [29]. In France in 1981, Levy found that PPV-23 was costeffective when applied to a population over 45 years [34]. Patrick and Woolley in 1981 showed the cost-benefits of the vaccine in people over 50 years and those with high-risk conditions (chronically cardiovascular, kidney, or liver ill, and diabetes), recommending the application in this population after to compare 3 alternatives: no vaccination, vaccinate all or vaccinate only high risk [35]. They considered a vaccine cost of US$8.61 per dose with polysaccharide vaccine of 14 serotypes plus operative costs, and only evaluate the pneumococcal pneumonia outcome. In 1997, the Advisory Committee on Immunization Practices (ACIP) updated the guidelines for recommending pneumococcal vaccine to vaccinate the entire population over 65 years, immunocompromised patients and high-risk populations [9].
In 2002, a cost-effectiveness study for people over 65 years in Catalonia (Spain) showed that pneumococcal vaccination for the population over 65 years was a cost-saving strategy to a cost per dose of D 11.51 [36]. Authors noted, that pneumococcal vaccination for older people was as cost-effective as other preventive interventions in this population, such as hypercholesterolemia, hypertension treatment, and smoking cessation therapies [36]. This study has limitations. There are no Colombian data for some of the epidemiological parameters related to the frequency of pneumococcal disease in adults above 60 years. Therefore, those indicators had to be estimated from the Latin American medical literature, which could potentially limit some of the conclusions. Despite that, the proportion of all pneumonias that could be associated to pneumococcal infection assumed in this analysis (10.5%) is a conservative estimate that may lead to a potential underestimation of the burden of disease and the benefits of pneumococcal vaccination [37–39]. Similarly population based studies from developed countries have found rates of all-cause pneumonia higher than 10% [40]. Also, important variables, such as the efficacy against all-cause pneumonia, incidence of IPD and case-fatality ratio are unavailable for Colombia. This is not a problem unique to our study. A review of the cost-effectiveness of pneumococcal vaccination strategies [29], highlighted the wide variation in assumptions for the key variables, even within population subgroups.
Fig. 3. Acceptability curve of cost-effectiveness analysis of PPV-23 vaccination in elderly Colombian population.
C. Casta˜ neda-Orjuela et al. / Vaccine 29 (2011) 7644–7650
Other limitations of our study are the non-inclusion of indirect effects of pneumococcal disease that underestimated the benefits of vaccination, and the non-consideration of antimicrobial resistance. Although the present study did not include otitis media as an outcome, the low frequency of this event in the elderly population may have a small effect in the results, underestimating the effect of vaccination. On the other hand bacteremia by pneumococcal also did not included which could lead to underestimate vaccination effects, but this should partially avoided when we included pneumonias and meningitis with blood isolation of S. pneumoniae. The efficacy of PPV-23 against all cause pneumonia was one of the most sensitive variables to changes. At efficacy values less than 4% PPV-23 is not cost-effective. The efficacy of PPV-23, for non-invasive pneumococcal pneumonia, has been always a highly controversial issue, especially in elderly population. The last WHO position paper on use of PPV-23 (2008) [41], recognizes that the evidence for efficacy and effectiveness in children and adults is poor, based on a recent review by Moberley and Huss in 2008 and 2009 respectively [17,42]. However it still acknowledges effectiveness from 50 to 80% for IPD prevention among adults [15]. In addition, the WHO paper highlights the existing challenges to estimate accurately the effectiveness of PPV-23 on the field, including the low frequency of IPD, the lack of consensus on diagnostic criteria for pneumococcal pneumonia, the variation of the effectiveness with age, and the severity of several underlying conditions associated with an increased risk of pneumococcal disease [41]. It is also recognized that very few randomized trials have been carried out and those who have done could have methodological problems that can potentially biased the results [15,43–45]. Despite the scarcity of evidence, WHO recommends PPV-23 vaccination in developed and developing countries given the high burden of pneumococcal disease among adults and children [41]. Despite of above descript limitations our analysis results are important as evidence for Colombian decision makers. However, other approach as budget impact analysis and political considerations can be important to evaluate in the Colombian and other developing countries settings. 5. Conclusion This analysis found that vaccination in the population over 60 year olds with the PPV-23 vaccine could be a very cost-effective strategy for the prevention of IPD in Colombia, a South American developing country. The results are consistent when the effect of non-invasive pneumococcal pneumonias is excluded. A surveillance analysis will be needed to evaluate the real impact of PPV-23 vaccination in the Colombian setting after its implementation. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.vaccine.2011.08.006. References [1] Fedson DS, Scott JA. The burden of pneumococcal disease among adults in developed and developing countries: what is and is not known. Vaccine 1999;17(July (Suppl. 1)):S11–8. [2] 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(April (13)):1729–35. [3] Ortqvist A, Hedlund J, Kalin M. Streptococcus pneumoniae: epidemiology, risk factors, and clinical features. Semin Respir Crit Care Med 2005;26(December (6)):563–74. [4] Monto AS. Acute respiratory infection in children of developing countries: challenge of the 1990s. Rev Infect Dis 1989;11(May–June (3)):498–505.
7649
[5] Burman LA, Norrby R, Trollfors B. Invasive pneumococcal infections: incidence, predisposing factors, and prognosis. Rev Infect Dis 1985;7(March–April (2)):133–42. [6] Marrie TJ, Durant H, Yates L. Community-acquired pneumonia requiring hospitalization: 5-year prospective study. Rev Infect Dis 1989;11(July–August (4)):586–99. [7] World Health Organization. Estimates of disease burden and costeffectiveness. Geneva; 2008 [cited 2010 junio]. Available from: http://www.who.int/immunization monitoring/burden/estimates burden/en/ index.html. [8] DANE. Cuadro 5. Defunciones, por grupos de edad y sexo, ˜ 2007. según lista de causas agrupadas 6/67 CIE-1 de OPS. Ano Bogota, D.C.: DANE; 2007 [cited 2011 May]. Available from: http://www.dane.gov.co/daneweb V09/index.php?option=com content&view =article&id=205&Itemid=119. [9] Prevention of pneumococcal disease: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep 1997;46(April (RR-8)):1–24. [10] Bogaert D, Hermans PW, Adrian PV, Rumke HC, de Groot R. Pneumococcal vaccines: an update on current strategies. Vaccine 2004;22(June (17–18)):2209–20. [11] NHS. In: NHS, editor. Pneumococcal vaccine for older people. London, UK: Department of Health; 2004. [12] Dhankhar P, Grabenstein JD, O’Brien MA, Dasbach EJ. Cost-effectiveness of stockpiling 23-valent pneumococcal polysaccharide vaccine to prevent secondary pneumococcal infections among a high-risk population in the United States during an influenza pandemic. Clin Ther 2010;32(August (8)):1501–16. [13] Watson L, Wilson BJ, Waugh N. Pneumococcal polysaccharide vaccine: a systematic review of clinical effectiveness in adults. Vaccine 2002;20(May (17–18)):2166–73. [14] Moore RA, Wiffen PJ, Lipsky BA. Are the pneumococcal polysaccharide vaccines effective? Meta-analysis of the prospective trials. BMC Fam Pract 2000;1:1. [15] Conaty S, Watson L, Dinnes J, Waugh N. The effectiveness of pneumococcal polysaccharide vaccines in adults: a systematic review of observational studies and comparison with results from randomised controlled trials. Vaccine 2004;22(August (23–24)):3214–24. [16] Melegaro A, Edmunds WJ. The 23-valent pneumococcal polysaccharide vaccine. Part I. Efficacy of PPV in the elderly: a comparison of meta-analyses. Eur J Epidemiol 2004;19(4):353–63. [17] Moberley SA, Holden J, Tatham DP, Andrews RM. Vaccines for preventing pneumococcal infection in adults. Cochrane Database Syst Rev 2008;1:CD000422. [18] De Graeve D, Lombaert G, Goossens H. Cost-effectiveness analysis of pneumococcal vaccination of adults and elderly persons in Belgium. Pharmacoeconomics 2000;17(June (6)):591–601. [19] Castaneda-Orjuela C, Alvis-Guzman N, de la Hoz-Restrepo F. The impact of pneumococcal disease on adults living in Bogota, Colombia, 2008. Rev Salud Publica (Bogota) 2010;12(February (1)):38–50. [20] DANE. Proyecciones de Población Municipales. Bogotá, D.C.: DANE; 2009 [2009 Feb 23]. Available from: http://www.dane.gov.co/index.php?option=com content&task=category§ionid=16&id=497&Itemid=995. [21] PROFAMILIA. Encuesta Nacional de Demografia y Salud Bogota D.C.; 2005. [22] Matute AJ, Brouwer WP, Hak E, Delgado E, Alonso E, Hoepelman IM. Aetiology and resistance patterns of community-acquired pneumonia in Leon, Nicaragua. Int J Antimicrob Agents 2006;28(November (5)):423–7. [23] Diaz A, Vera C, Santamarina J, Izarduy L, Bagnesco R, Grinspon S, et al. Neumonía adquirida en la comunidad en acncianos que requieren internación.Cuadro clínico y pronóstico. Medicina (Arg) 1999;59(6):731–8. [24] Szot J. Mortalidad por enfermedades respiratorias en Chile durante 1999. Rev Chil Enferm Respir 2003;19(1):8–14. [25] Dickinson FO, Pérez AE. Meningoencefalitis bacterianas en Cuba. Rev Cubana Hig Epidemiol 2001;39(2):86–94. [26] Dickinson Meneses F, Pérez Rodríguez A. Meningoencefalitis bacterianas en el adulto mayor cubano 1998–2000. Geriátrika 2002;18(6):32–40. [27] Gomes I, Lucena R, Melo A. Clinical and laboratory characteristics of pyogenic meningitis in adults. Arq Neuropsiquiatr 1997;55(September (3B)): 584–7. [28] Rossoni AM, Dalla Costa LM, Berto DB, Farah SS, Gelain M, Brandileone MC, et al. Acute bacterial meningitis caused by Streptococcus pneumoniae resistant to the antimicrobian agents and their serotypes. Arq Neuropsiquiatr 2008;66(September (3A)):509–15. [29] Ogilvie I, Khoury AE, Cui Y, Dasbach E, Grabenstein JD, Goetghebeur M. Cost-effectiveness of pneumococcal polysaccharide vaccination in adults: a systematic review of conclusions and assumptions. Vaccine 2009;27(August (36)):4891–904. [30] WHO Commission on Macroeconomics and Health. Macroeconomics and health: investing in health for economic development. Report of the Commission on Macroeconomics and Health. Geneva; 2001. [31] Willems JS. Cost effectiveness and cost benefit of vaccination against pneumococcal pneumonia: implications for clinical practice. Infect Control 1982;3(July–August (4)):299–302. [32] Willems JS, Sanders CR, Riddiough MA, Bell JC. Cost effectiveness of vaccination against pneumococcal pneumonia. N Engl J Med 1980;303(September (10)):553–9. [33] Ament A, Baltussen R, Duru G, Rigaud-Bully C, de Graeve D, Ortqvist A, et al. Cost-effectiveness of pneumococcal vaccination of older people: a study in 5 western European countries. Clin Infect Dis 2000;31(August (2)):444–50.
7650
C. Casta˜ neda-Orjuela et al. / Vaccine 29 (2011) 7644–7650
[34] Levy E. Cost effectiveness of pneumococcal vaccine (author’s transl). Rev Epidemiol Sante Publique 1981;29(2):133–53. [35] Patrick KM, Woolley FR. A cost–benefit analysis of immunization for pneumococcal pneumonia. JAMA 1981;245(February (5)):473–7. [36] Plans P. Cost-effectiveness of 23-valent antipneumococcical vaccination in Catalonia (Spain). Gac Sanit 2002;16(September–October (5)):392–400. [37] Ruiz-Gonzalez A, Falguera M, Nogues A, Rubio-Caballero M. Is Streptococcus pneumoniae the leading cause of pneumonia of unknown etiology? A microbiologic study of lung aspirates in consecutive patients with community-acquired pneumonia. Am J Med 1999;106(April (4)):385–90. [38] Mandell LA, Wunderink RG, Anzueto A, Bartlett JG, Campbell GD, Dean NC, et al. Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community-acquired pneumonia in adults. Clin Infect Dis 2007;44(March (Suppl. 2)):S27–72. [39] Brown PD, Lerner SA. Community-acquired pneumonia. Lancet 1998;352(October (9136)):1295–302. [40] Jackson ML, Neuzil KM, Thompson WW, Shay DK, Yu O, Hanson CA, et al. The burden of community-acquired pneumonia in seniors: results of a populationbased study. Clin Infect Dis 2004;39(December (11)):1642–50.
[41] World Health Organization, 23-valent pneumococcal polysaccharide vaccine. WHO position paper. Wkly Epidemiol Rec 2008;83(October (42)): 373–84. [42] Huss A, Pippa Scott P, Stuck AE, Trotter C, Egger M. Efficacy of pneumococcal vaccination in adults: a meta-analysis. Can Med Assoc J 2009;180(January (1)):48–58. [43] Butler JC, Breiman RF, Campbell JF, Lipman HB, Broome CV, Facklam RR. Pneumococcal polysaccharide vaccine efficacy. An evaluation of current recommendations. JAMA 1993;270(October (15)):1826– 31. [44] Mangtani P, Cutts F, Hall AJ. Efficacy of polysaccharide pneumococcal vaccine in adults in more developed countries: the state of the evidence. Lancet Infect Dis 2003;3(February (2)):71–8. [45] Zhogolev SD, Ogarkov PI, Mel’nichenko PI. The prophylaxis of nonhospital pneumonia using 23-valent pneumococcus vaccine in the military collectives. Voen Med Zh 2004;325(December (12)):35–43, 96. [46] Standaert B, Parez N, Tehard B, Colin X, Detournay B. Cost-effectiveness analysis of vaccination against rotavirus with RIX4414 in France. Appl Health Economics Health Policy 2008;6(4):199–216.