Vaccination of children with a live-attenuated, intranasal influenza vaccine - analysis and evaluation through a Health Technology Assessment.

BACKGROUND: Influenza is a worldwide prevalent infectious disease of the respiratory tract annually causing high morbidity and mortality in Germany. Influenza is preventable by vaccination and this vaccination is so far recommended by the The German Standing Committee on Vaccination (STIKO) as a standard vaccination for people from the age of 60 onwards. Up to date a parenterally administered trivalent inactivated vaccine (TIV) has been in use almost exclusively. Since 2011 however a live-attenuated vaccine (LAIV) has been approved additionally. Consecutively, since 2013 the STIKO recommends LAIV (besides TIV) for children from 2 to 17 years of age, within the scope of vaccination by specified indications. LAIV should be preferred administered in children from 2 to 6 of age. The objective of this Health Technology Assessment (HTA) is to address various research issues regarding the vaccination of children with LAIV. The analysis was performed from a medical, epidemiological and health economic perspective, as well as from an ethical, social and legal point of view.
METHOD: An extensive systematic database research was performed to obtain relevant information. In addition a supplementary research by hand was done. Identified literature was screened in two passes by two independent reviewers using predefined inclusion and exclusion criteria. Included literature was evaluated in full-text using acknowledged standards. Studies were graded with the highest level of evidence (1++), if they met the criteria of European Medicines Agency (EMA)-Guidance: Points to consider on applications with 1. meta-analyses; 2. one pivotal study.
RESULTS: For the medical section, the age of the study participants ranges from 6 months to 17 years. Regarding study efficacy, in children aged 6 months to ≤7 years, LAIV is superior to placebo as well as to a vac-cination with TIV (Relative Risk Reduction - RRR - of laboratory confirmed influenza infection approx. 80% and 50%, respectively). In children aged >7 to 17 years (= 18th year of their lives), LAIV is superior to a vaccination with TIV (RRR 32%). For this age group, no studies that compared LAIV with placebo were identified. It can be concluded that there is high evidence for superior efficacy of LAIV (compared to placebo or TIV) among children aged 6 months to ≤7 years. For children from >7 to 17 years, there is moderate evidence for superiority of LAIV for children with asthma, while direct evidence for children from the general population is lacking for this age group. Due to the efficacy of LAIV in children aged 6 months to ≤7 years (high evidence) and the efficacy of LAIV in children with asthma aged >7 to 17 years (moderate evidence), LAIV is also very likely to be efficacious among children in the general population aged >7 to 17 years (indirect evidence). In the included studies with children aged 2 to 17 years, LAIV was safe and well-tolerated; while in younger children LAIV may increase the risk of obstruction of the airways (e.g. wheezing). In the majority of the evaluated epidemiological studies, LAIV proved to be effective in the prevention of influenza among children aged 2-17 years under everyday conditions (effectiveness). The trend appears to indicate that LAIV is more effective than TIV, although this can only be based on limited evidence for methodological reasons (observational studies). In addition to a direct protective effect for vaccinated children themselves, indirect protective ("herd protection") effects were reported among non-vaccinated elderly population groups, even at relatively low vaccination coverage of children. With regard to safety, LAIV generally can be considered equivalent to TIV. This also applies to the use among children with mild chronically obstructive conditions, from whom LAIV therefore does not have to be withheld. In all included epidemiological studies, there was some risk of bias identified, e.g. due to residual confounding or other methodology-related sources of error. In the evaluated studies, both the vaccination of children with previous illnesses and the routine vaccination of (healthy) children frequently involve cost savings. This is especially the case if one includes indirect costs from a societal perspective. From a payer perspective, a routine vaccination of children is often regarded as a highly cost-effective intervention. However, not all of the studies arrive at consistent results. In isolated cases, relatively high levels of cost-effectiveness are reported that make it difficult to perform a conclusive assessment from an economic perspective. Based on the included studies, it is not possible to make a clear statement about the budget impact of using LAIV. None of the evaluated studies provides results for the context of the German healthcare setting. The efficacy of the vaccine, physicians' recommendations, and a potential reduction in influenza symptoms appear to play a role in the vaccination decision taken by parents/custodians on behalf of their children. Major barriers to the utilization of influenza vaccination services are a low level of perception and an underestimation of the disease risk, reservations concerning the safety and efficacy of the vaccine, and potential side effects of the vaccine. For some of the parents surveyed, the question as to whether the vaccine is administered as an injection or nasal spray might also be important.
CONCLUSION: In children aged 2 to 17 years, the use of LAIV can lead to a reduction of the number of influenza cases and the associated burden of disease. In addition, indirect preventive effects may be expected, especially among elderly age groups. Currently there are no data available for the German healthcare setting. Long-term direct and indirect effectiveness and safety should be supported by surveillance programs with a broader use of LAIV. Since there is no general model available for the German healthcare setting, statements concerning the cost-effectiveness can be made only with precaution. Beside this there is a need to conduct health eco-nomic studies to show the impact of influenza vaccination for children in Germany. Such studies should be based on a dynamic transmission model. Only these models are able to include the indirect protective effects of vaccination correctly. With regard to ethical, social and legal aspects, physicians should discuss with parents the motivations for vaccinating their children and upcoming barriers in order to achieve broader vaccination coverage. The present HTA provides an extensive basis for further scientific approaches and pending decisions relating to health policy.

Background

Influenza is a worldwide prevalent infectious disease of the respiratory tract resulting from influenza viruses A or B, causing high morbidity and mortality in Germany. The Robert Koch Institute (RKI) estimates the disease burden of influenza in Germany for the population as a whole during the season of 2012/13 as follows [1]:

Number of additional GP visits (GP = general practitioner) during the influenza epidemic (excess consultations): 7.7 million (95% CI 7.1–8.1 million).

Influenza-related inability to work or need for care in dependent children and non-workers: 4.3 million (95% CI 4.0–4.7 million).

Number of influenza-related additional hospital admissions: 32,000 (95% CI: 28,000–35,000).

Children up to four years of age caused the highest number of consultations due to acute respiratory tract infections. For this age group, the RKI estimates the number of additional GP visits as 28,000 per 100,000 children. The number of babies and infants in need of care due to influenza was estimated as 270,000 (95% CI 210,000–330,000). The highest incidence of consultations was estimated for school children (five to 14 years). For this age group, the number of additional GP visits was estimated as 16,000 per 100,000 children. 470,000 children between the ages of five to 14 years (95% CI 410,000–530,000) were absent from school due to influenza [1].

Influenza is preventable by vaccination; the influenza vaccination is recommended by the German Standing Committee on Vaccinations (STIKO) as a standard vaccination for the elderly over the age of 60. To date, parentally administered trivalent inactivated vaccines (TIV) have been used almost exclusively, even if in 2011 nasally administered live attenuated inactive vaccine (LAIV) has also been licensed. In August 2013, STIKO amended its vaccination recommendations for influenza noting that children and adolescents between two and 17 years of age could be vaccinated with a TIV or LAIV, provided there were no contraindications according to the prescribing information. Preferred use of LAIV vaccination for children aged two to six years is recommended.

After an influenza infection, there is generally protection against reinfection with this specific influenza virus (“immunological memory”). However, because of the constant mutation of circulating influenza viruses, this does not lead to reliable protection against annual or seasonally reoccurring infections with shift variants – and the same also applies to vaccine protection. For that reason, vaccination against influenza should be repeated annually.

In terms of their own disease burden as well as their role in the dissemination of the disease to other age groups, children are considered as a particularly important target group for vaccination against influenza. Because influenza epidemics do not spread across all age groups at the same speed, but primarily affect children, a general vaccination of children may have a beneficial effect in terms of “herd protection”.

Influenza and its consequences are thus of great significance to the public health system because of increased demands on health services with associated costs for the German health care system. However, cost-of-illness analyses for influenza in Germany are difficult to conduct due to significant annual/seasonal variations in spread and disease burden. Reducing the influenza-related disease burden is an important objective of public health policy. In this regard, high effectiveness of available vaccines as well as the achievement of high vaccination coverage rates in relevant target groups could be of crucial importance for public health policy.

Therefore, the aim of this study was to answer various research questions on the vaccination of children with LAIV – from a medical, epidemiological and health-economic perspective as well as from an ethical, social and legal (ESL) one.

Methodology

An extensive systematic database research was performed to obtain relevant information. In addition, a supplementary hand search was carried out. A search strategy as well as inclusion or exclusion criteria in accordance with German Institute of Medical Documentation and Information (DIMDI) specifications were developed for all areas (medical, epidemiological and health economical as well as ethical, social and legal aspects). The search terms or their combinations were prepared in detail over several trial runs (see Attachment 1). No search for “grey literature” was conducted.

Titles and abstracts of identified publications were first checked on the basis of specified inclusion or exclusion criteria and the remaining sources then assessed on the basis of their full-text versions. This review was conducted by two independent experts. Criteria for the literature selection were defined in accordance with the PICO (Population-Intervention-Comparison-Outcome) framework [2] and supplemented with further specifications regarding study type and setting (Attachment 2). The content of sources remaining after this process were evaluated in a structured form and assessed with the aid of recognised standards for the assessment of scientific studies. The flow diagram of the search is shown in Attachment 3 .

For the quality assessment of the included literature, medical and epidemiological studies were assessed on the basis of checklists provided by the German Scientific Working Group Technology Assessment for Health Care [3]. In addition, in order to evaluate the evidence in medical and epidemiological publications, the classification developed by the Scottish Intercollegiate Guidelines Network (SIGN) Grading Review Group was used [4]. Included reports on studies were classified in accordance with their evidence grade. Primary studies were only graded with the highest level of evidence (1++) if they met the strict criteria for a pivotal study as defined by the European Medicines Agency (EMA-)-Guidance Points to consider on applications with 1. meta-analyses; 2. one pivotal study [5]. Assessment of the methodological quality of included health-economic studies was guided by the Good Practice Guidelines for Decision-Analytic Modelling in Health Technology Assessment [6] as well as the Criteria list for assessment of methodological quality of economic evaluations [7].

This was followed by comprehensive documentation of the methodological aspects or the results from the included literature, respectively. Results were exhibited in a standardised format in order to permit an assessment in excerpts as required for the HTA (Health Technology Assessment) report on the review and comparability of studies. The remaining literature was used to provide answers to the following research questions:

Medical research questions

What is the efficacy of LAIV vaccination in children and adolescents up to the age of 18 years compared to other interventions and placebo?

How safe is LAIV for children and adolescents up to the age of 18 years in comparison to other interventions and placebo?

Are there differences regarding the efficacy and safety of LAIV in relation to particular factors?

From a medical perspective, where is further research required due to inconsistent study results or missing studies?

In order to answer questions regarding the efficacy of LAIV, the primary emphasis was placed on the endpoint of influenza cases confirmed through laboratory analysis. Additionally, further direct patient relevant endpoints, such as otitis media, were depicted.

Epidemiological research questions

How effective is LAIV in children and adolescents up to the age of 18 years in comparison to other interventions, placebo or no intervention (effectiveness)?

What is the safety of LAIV in children and adolescents up to the age of 18 years in comparison to other interventions, placebo or no intervention?

Does the vaccination against influenza of children and adolescents up to the age of 18 years produce indirect protective effects (“herd protection”)?

Are there differences regarding the effectiveness and safety of LAIV in relation to particular factors?

Health-economic research questions

What is the general assessment of the cost-effectiveness of vaccination against influenza in children and adolescents up to the age of 18 years (efficiency)?

Is the use of LAIV in children and adolescents up to the age of 18 years a cost-effective option?

What are the budgetary impacts of using LAIV?

From a health-economic perspective, where is further research required due to inconsistent study results or missing studies? What methodological recommendations can be given for future studies?

Ethical/social/legal research questions

Which factors influence individual decision-making in children and adolescents up to the age of 18 years for or against an influenza vaccination? In this decision, what is the impact of administration method and information provided by GP? What are the preferences of children and adolescents up to the age of 18 years, their parents or guardians, as well as society regarding the vaccination against influenza?

Do ethical conflicts exist between the right to self-determination by children and parents and the societal desire for general vaccine protection (herd immunity)? Is it permissible to restrict children and adolescents, as spreaders of infection, or their parents or guardians in their individual right to decide for or against vaccination in order to protect vulnerable groups from infection?

What is the current situation in Germany regarding access to vaccination against influenza in general or specifically for different vaccines? What is the assessment of the practice of reimbursement? Which access barriers exist? What is the impact of the current practice of a tendering process for influenza vaccines by the statutory health insurance?

Which legal aspects should be taken into account?

Results of the literature search

On the basis of the search strategy, the systematic literature search identified the following: 375 medical hits, 1,228 hits in the field of epidemiology, 474 economic publications and 411 hits for ESL (review of search process and results in Attachment 3 and Attachment 4).

A total of 37 studies were included in the assessment of the medical section (see Table A4-1, Table A4-2 and Table A4-3 in Attachment 4), 16 studies in the assessment of the epidemiological section (see Table A4-4, Table A4-5 and Table A4-6 in Attachment 4), 27 studies in the assessment of the health-economic section (see Table A4-7, Table A4-8 and Table A4-9 in Attachment 4) and seven studies in the assessment of the ESL section [8], [9], [10], [11], [12], [13], [14].

Answering the research questions

What is the efficacy of the LAIV vaccination in children and adolescents up to the age of 18 years compared to other interventions and placebo?

This HTA determines efficacy primarily in terms of the prevention of laboratory-confirmed cases of symptomatic influenza. Randomized Controlled Trials (RCT) reveal that for children from six months to 18 years of age, LAIV is superior to placebo and TIV in terms of efficacy. No efficacy studies were identified for children aged <6 months.

For the differentiated characterisation of available evidence, characteristics of the target population (general population versus children with asthma) as well as the age group have to be considered (Table 1 (Tab. 1)):

Table 1

Scope of evidence on the efficacy of LAIV (in comparison to placebo or TIV) derived from RCT, measured in terms of the prevention of laboratory-confirmed cases of influenza

For children from six months to ≤7 years, both LAIV and placebo are superior to TIV. Efficacy compared to placebo or to TIV for laboratory-confirmed influenza was approx. 80% or 50%, respectively. The evidence is high for children from the general population (i.e. concomitant diseases were no inclusion criterion in these studies) and also for children with asthma.

For children aged >7 to 17 years (= 18th birthday), LAIV is superior to vaccinations with TIV. No studies are available that compared LAIV with placebo. Efficacy as compared to TIV was approx. 32%. The evidence for this statement originates from one randomised open study, conducted during one influenza season on children with asthma [15]. For that reason, the evidence for children with asthma has to be rated moderate; no studies were available for children in the general population. In view of the proven efficacy of LAIV in children aged six months to ≤7 years (high evidence) and the proven efficacy of LAIV in children with asthma aged >7 to 17 years (moderate evidence), LAIV is also very likely to be efficacious in children from the general population aged >7 to 17 years. However, because this is indirect evidence, the evidence for this statement has to be rated as low (Table 1 (Tab. 1), [16], [17]).

How safe is LAIV for children and adolescents up to the age of 18 years compared to other interventions and placebo?

In the included studies with children aged two to 17 years, LAIV was safe and well-tolerated. The most common side effects were local reactions to the vaccine (rhinorrhoea/rhinitis/blocked nose) as well as general signs of reactogenicity (e.g. fever, loss of appetite, vomiting, reduced activity, and increased irritability). In the included studies, there were no differences for this age group in the frequency of (severe) adverse events that led to the termination of treatment (as far as reported). There was no report of fatal events potentially related to vaccination in the included studies.

One study [18] indicates a higher rate of hospital admissions for a variety of reasons as well as a higher rate of severe adverse events associated with LAIV in children aged six to eleven months as compared to TIV. The same study also indicates a higher incidence rate of acute wheezing in children aged six to 24 months under LAIV. Based on these findings, the use of LAIV was restricted to children from 24 months to 18 years of age.

Are there differences regarding the efficacy and safety of LAIV in relation to particular factors?

Gender: There are indications from a meta-analysis that the efficacy of LAIV may be higher in females receiving the vaccine than in males; however, the superiority of LAIV (compared to TIV and placebo) was observed for both genders (quantitative interaction).

Age: A subgroup analysis of four studies [19] showed LAIV to be consistently effective in children from six months to six years of age across the analysed age range. No studies were identified that analysed whether the efficacy in the age range <6 years differed from that in the age range from seven to 17 years. With regard to safety endpoints, two studies indicated that the tolerability of LAIV declines in younger children, particularly less than 24 months of age, compared to older children (higher rate of acute wheezing/reactive respiratory ailments and hospital admissions for any reason).

Region: There are indications that the efficacy of LAIV is lower in studies conducted in Asia than in other regions.

From a medical perspective, where is further research required due to inconsistent study results or missing studies?

Evidence on the efficacy of LAIV for children aged from seven to 17 years is limited (see answer to the first research question). In this age population, additional studies, particularly with children from the general population (without concomitant diseases as an inclusion criterion), would be desirable. A particular focus of such studies could be the analysis of whether superiority over comparators can be concluded for the entire age range (from seven to 17 years) or whether there are indications of age-related interactions.

Subgroup analyses of the studies by Ashkenazi et al. [16] and Belshe et al. [18] indicate that LAIV was safe for children ≥24 months with a medical history of mild or moderate asthma or wheezing, specifically in relation to acute wheezing and hospital admissions for any reason. However, there are no studies to date on children with acute wheezing or children with severe asthma; additional studies would be desirable for these risk populations. If the tolerability of the vaccination in children aged twelve to 23 months can be improved by using alternative vaccination schedules (e.g. ascending doses; initial vaccination with TIV, followed by LAIV) might also be investigated in additional studies.

A tetra- or quadrivalent LAIV (LAIV-Q) has recently been licensed in Germany. This is in accordance with the amended WHO (World Health Organisation) recommendation for the composition of influenza vaccines. From the vaccination season 2014-15, LAIV-Q will completely replace the current trivalent live vaccine (LAIV-T). LAIV-Q is intended to ensure protection against both of the influenza-B strains currently in global circulation (Yamagata and Victoria) and could thus show a greater efficacy compared to placebo treatments. More detailed studies on the efficacy and effectiveness of this vaccine would be desirable.

How effective is LAIV in children and adolescents up to the age of 18 years compared to other interventions, placebo or no intervention?

The great majority of evaluated studies showed that LAIV was effective in the prevention of influenza in clinical practice under real world conditions. Compared to non-vaccination, the effect size can be estimated approx. 10%. The interpretation of these results must take into consideration that acute respiratory tract infection was the chosen target parameter; however, this could not just be caused by influenza but also by a number of other pathogens. The superiority of LAIV over TIV as shown in the included observational studies has to be interpreted with caution due to methodological reasons (high risk of bias, e.g. as a consequence of limited consideration of potential confounders). Except of age, the analyses did not consider any other potential confounder such as socio-economic variables, day-care attendance, domestic living conditions and/or the number of persons per household.

What is the safety of LAIV in children and adolescents up to the age of 18 years compared to other interventions, placebo or no intervention?

In the included observational studies, LAIV was safe and well tolerated. There were no indicators for an increased rate of adverse effects that may have challenged the positive risk-benefit ratio. Differences between LAIV and TIV as shown in the included observational studies have to be interpreted with caution for methodological reasons. In particular, this applies to parameters used in the assessment of undesirable effects because of their partial overlap with symptoms of respiratory infections.

There were no indicators of clinically relevant higher risks for undesirable effects under LAIV compared to TIV. This also applies to the administration to children with mild chronic-obstructive diseases. There was a high risk of bias in the studies, e.g. as a consequence of limited consideration of potential confounders.

Does the vaccination against influenza of children and adolescents up to the age of 18 years produce indirect protective effects (“herd protection”)?

Included studies investigating these questions showed a herd protection effect for non-vaccinated elderly population groups, based on decreased rates of Medically Attended Acute Respiratory Illness (MAARI). This protection was already evident at rather low vaccination rates of children within the respective vaccination programmes compared to a potentially more comprehensive vaccination rate as part of a general vaccination recommendation for children. However, it seems very difficult to quantify this effect in a valid way because too many different influencing factors (including age structure, frequency of nursery or school attendance, contact rates between age groups) would need to be considered. There was a high risk of bias in the studies, e.g. in consequence of insufficient consideration given to disturbance variables. With the exception of age, the analyses did not consider any other potential confounders of the association under study (such as socio-economic variables, day-care attendance, domestic living conditions and the number of persons per household).

Are there differences regarding the effectiveness and safety of LAIV in relation to particular factors?

Effectiveness: The evaluated studies indicate that the vaccination is most effective in pre-school children and that this effect is continually decreasing with increasing patient age. No differences in effectiveness were found in respect of other factors such as gender, ethnic origins or state of health/comorbidities (unless the applicability of LAIV was ruled out as a result).

Safety: Even though there is some differentiation according to age groups in the included studies, it is very difficult to deduce age-related differences in relation to undesirable effects under real world conditions because of the original intention of the studies themselves, i. e. compare LAIV to TIV, or the manner in which the results are presented. Furthermore, there were no indications that the aforementioned factors had an impact on the safety of LAIV.

What is the general assessment of the cost-effectiveness of vaccination against influenza of children and adolescents up to the age of 18 years?

In the evaluated studies, both the vaccination of children with previous illnesses and the routine vaccination of (healthy) children are frequently associated with cost savings. This is especially the case if indirect costs from a societal perspective are also included. From a payer's perspective, routine vaccination of children is often regarded as a highly cost-effective form of intervention. However, not all studies give consistent results. There are individual reports of comparatively high cost-benefit ratios that complicate a final assessment from an economic perspective.

Is the use of LAIV in children and adolescents up to the age of 18 years a cost-effective option?

A number of studies assess the use of LAIV as a cost-saving or cost-effective intervention [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47]. However, not all studies reach consistent results. Compared to TIV, vaccination with LAIV represents the more cost-effective alternative. One of the assessed studies evaluated both the use of LAIV and TIV and assumed an identical price of LAIV and TIV based on the price level of the latter.

What are the budgetary impacts of using LAIV?

Based on the included studies, it is not possible to make a clear statement about the budgetary impacts of using LAIV.

From a health-economic perspective, where is further research required due to inconsistent study results or missing studies? What methodological recommendations can be given for future studies?

None of the assessed health-economic studies provides results based on the German health care context. The transfer of results from other countries to a national context is viewed as problematic due to varying health care structures, among other reasons. For that reason, there is a growing need for health-economic studies on the effects of influenza vaccination of children that give appropriate consideration to the German health care context. Such studies should be based on a dynamic transmission model which includes indirect protective effects of vaccination.

Ethical, social and legal research questions

Which factors influence individual decision-making in children and adolescents up to the age of 18 years for or against an influenza vaccination? In this decision, what is the impact of the administration method and information provided by GP? What are the preferences of children and adolescents up to the age of 18 years, their parents or guardians, as well as society regarding the vaccination against influenza?

The efficacy of the vaccine, physician recommendations as well as a potential reduction in influenza symptoms seemed to have played a role in the parent and guardian decision-making for their children. Major barriers to the utilisation of influenza vaccination services include low level of perception and underestimation of disease risk, reservations concerning the safety and efficacy of the vaccine as well as potential side effects of the vaccine. For some of the parents surveyed, the question as to whether the vaccine is administered as injection or nasal spray might also be important. Efficacy and the method of administration could be the most important features in decision-making regarding childhood vaccination. For children, a preference for LAIV over TIV could be assumed. In this particular respect, however, the significance of the evidence is very limited; to permit further conclusions, studies should be conducted within the German health care context. On the whole, it is evident that the attitude of the GP has a significant impact on the decision by parents and guardians in favour of vaccination. GP are the most important source of information about influenza vaccines. For that reason, physician recommendations play an important role in the decision-making process. However, other professions involved in the care of children, e.g. nursery staff, communicate their individual views on vaccination to parents. The availability of such information is an important influencing factor for the decision in favour of influenza vaccinations. Information about disease risk, efficacy, safety, possible side effects, but also about positive effects for society via herd protection is of great relevance. Parents and guardians, children and adolescents as well as service providers are to receive comprehensive information via a range of different media, with information material provided.

Do ethical conflicts exist between the right to self-determination by children and parents and societal desires for general vaccine protection (herd immunity)? Is it permissible to restrict children and adolescents, as spreaders of infection, or respectively their parents or guardians in their individual right to decide for or against vaccination in order to protect vulnerable groups from infection?

Ethical conflicts between children’s or parents’ right to self-determination are possible. For that reason, careful consideration has to be given to the question as to whether children and adolescents could also be vaccinated against their parents’ will. Impairments of the parents’ right to a decision may be more advisable in line with the level of risk to the child's welfare. A similar approach is also appropriate for the question as to whether or not vaccinations can be ordered against the will of an individual for the benefit of society or the protection of vulnerable populations. A definite answer to this problem cannot be given as a social debate is necessary. However, vaccination against influenza as compulsory measure should be viewed and discussed most critically. Moreover, vaccination information should be provided and discussed, vaccination characteristics should be further explained, and methods of vaccine administration should be taken into consideration as well as examining how public opinion is influenced.

What is the current situation in Germany regarding access to vaccination against influenza in general or specifically for different vaccines? What is the assessment of the practice of reimbursement? Which access barriers exist? What impact of the current practice of a tendering process for influenza vaccines by the statutory health insurance?

The current situation of influenza vaccine access in Germany is to be analysed critically. Particularities of reimbursement policies of the various health insurance companies should have no impact on the reimbursement of a vaccination generally recommended by the STIKO. In addition, the reference pricing system is to be analysed critically due to the risk of competitive distortion. In addition to the reference pricing system, the discount system comprehensively used by individual health insurance companies has to be questioned because of the conflicting objectives between cost savings for the insured and ensuring the supply of influenza vaccines. Bottlenecks in supply, which are partially caused by this, have a negative impact on proposed high vaccinations rates, as the desired target in terms of health policy. Furthermore, an exclusive focus on price as the criterion for decisions and disregarding further aspects relating to quality and economy is counter-productive. If certain vaccines show a superior protective effect in individual target groups, access via the reimbursement of costs should be ensured. As STIKO recommends LAIV as preferred option for children aged two to six years, access to LAIV for this target group should be made possible for all statutory health insurance members. In the future, there should be a fundamental debate whether the extra benefit of a particular vaccine could not also be reflected in a higher market price.

Which legal aspects should be taken into account?

Because of the current STIKO recommendations, legal aspects concerning information and liability play a minor role. No further important legal aspects have been raised so far. However, it is to be noted that medical service providers to date are encouraged to meet particularly high standards in informing patients and parents in the absence of an indication for vaccination (according to STIKO recommendation) for a particular patient group. Accordingly, questions of liability could become more important.

Discussion

Study design of clinical and epidemiological studies

The clinical studies that were included to answer the medical questions had a high to very high internal validity. The primary efficacy endpoint measured in efficacy studies (influenza infection confirmed in laboratory tests by polymerase chain reaction (PCR) or virus culture) was of a high methodological standard. The inclusion and exclusion criteria applied were appropriate and did not impede the transferability of the clinical studies to the reality of medical practice. In three of the efficacy studies conducted over two influenza seasons [48], [49], [50], participants did not undergo randomisation in the second vaccination season. This is potentially problematic as participation in the second study could have been influenced by effects of the vaccination in the first study, such as side effects or incidence of influenza infections during the first season. Thus, the equal distribution of characteristics of study participants achieved by randomisation can no longer be guaranteed. The other important methodological limitation of included RCTs is the lack of blinding in some studies.

As expected, study types in the epidemiological section were less consistent, e.g. in the collection of primary data on the one hand and the use of secondary data e.g. compiled by health insurances on the other. This inconsistency in epidemiological designs is somewhat counterbalanced by the fact that these studies were generally conducted in everyday medical care environments, thus meeting the transferability and wider applicability criterion of an intervention as desired for an HTA. It is to be assumed that vaccination effects were increasingly higher according to when the respective influenza outbreak began after the period of vaccination; this respective time interval could represent a potential distortion factor.

Age groups in clinical studies

In the interpretation of clinical efficacy results, two aspects are of special importance: the age of children included in the study and the characteristics of the target population (particularly “healthy” children versus children with relevant morbidity). Most of the included clinical studies had strict age range specifications for the inclusion of participants. Because of indications of an age-specific effect and side-effect spectrum of LAIV, results from those studies cannot simply be transferred to children of different ages. In contrast to epidemiological studies that mainly investigated vaccination effects in influenza seasons for healthy children predominantly between 18 months and 17 years of age, the body of evidence particularly in respect to the (clinical) efficacy of LAIV in children from seven to 17 years of age is limited. Additional studies, especially with children from the general population (i.e. without concomitant diseases as an inclusion criterion) would be desirable. Investigating the potential interaction between the efficacy of LAIV and age could be a potential focus. Subgroup analyses indicated that LAIV was safe for children aged ≥24 months with a medical history of mild or moderate asthma or wheezing, specifically with respect to acute wheezing and hospital admissions for any reason. However, there no studies on children with a history of acute wheezing or children with severe asthma were identified; additional studies would be desirable for these ‘at risk’ populations. Additional studies could also investigate whether in relation to incidences of acute wheezing the tolerability of LAIV could be improved with the aid of different vaccination schedules (e.g. ascending doses; initial vaccination with TIV, followed by LAIV).

Transferability of study results

It is questionable to which degree efficacy/efficiency and safety of LAIV in the various infantile age groups are comparable to each other because some studies have a comparatively narrow age range, others a broad one.

No data is available to date for the German health care context. Most studies were conducted in the United States with the majority of the epidemiological studies being conducted in different settings of general health care. The epidemiological studies cover the influenza seasons from 1998/99 to 2009/10 and generally recorded healthy children, but also some children with chronic respiratory tract diseases and in some instances even included further household members. These studies investigated LAIV either in comparison to TIV or to non-vaccination. The general target parameter for effectiveness was the prevention of acute respiratory tract infections or acute respiratory tract infections requiring medical treatment. In terms of safety, adverse events or severe adverse events both requiring medical treatment were included.

For that reason, the long-term efficacy/effectiveness and safety of LAIV should be accompanied and investigated in additional evaluation programmes.

Impact of the herd protection effect

There is general agreement in the analysed literature regarding the targeted vaccination against influenza in children and its ability to make a significant contribution towards reducing the disease burden (reduction of MAARI rates) in higher age groups. Apart from age, the analyses did not consider other potentially confounding factors a fact, which leaves the possibility of bias due to residual confounding.

Cost-effectiveness and access to vaccination services

The great majority of health-economic studies assess the influenza vaccination of children as a cost-saving and highly cost-effective investment. Particularly, relevant factors of influence are the choice of perspective, the price of the vaccination and the consideration of indirect protective effects. No health-economic model is available for the German health care context; this makes the transferability of the results more difficult. For that reason, there is need for health-economic studies on the effects of influenza vaccination of children in Germany. Such a study should be based on a dynamic transmission model to include the indirect protective effects of vaccination.

The current situation of access to influenza vaccines in Germany is to be analysed critically. Peculiarities of individual health insurance companies should have no impact on the reimbursement of vaccinations as recommended by the STIKO. In addition to the reference pricing system, the discount system comprehensively used by individual health insurance companies has to be questioned because of the conflicting objectives between cost savings for the insured and ensuring the supply of influenza vaccines. Bottlenecks in supply partially caused by this have a negative impact on reaching high vaccinations rates, a desired health policy goal. Furthermore, an exclusive focus on price as the criterion for decisions and a disregard of other aspects relating to quality and economy is not very productive. If certain vaccines show a superior protective effect in individual target groups, increased access via the reimbursement of costs should be ensured. As STIKO recommends LAIV as the preferred option for children aged two to six years, general access to LAIV for this target group should be made possible for all statutory health insurance patients. The German Ethics Council is currently discussing the ethical dimensions of compulsory vaccinations [51].

Methodology

Literature search

Even though all search terms and their combinations were developed in detail in several runs, it is possible that some relevant search terms were not included reducing the sensitivity of the search. There could also be some language bias as publications in languages other than English or other global languages may not have been displayed in the searched databases, or may not have included an English language abstract as well as publication bias due to preference given to result-dependent publications, specifically to articles showing positive or “significant” results of innovations.

Literature selection

The first selection step on the basis of title and abstract naturally opens up the possibility of impaired sensitivity through the possibility of not identifying relevant articles as such. Despite the use of two experts, misclassifications are theoretically also possible during the subsequent full-text selection which may have led to the inclusion or exclusion of individual studies. In this case, the compilers of a systematic review face a real dilemma: on the one hand, the aim must be to identify as many relevant publications as possible; on the other hand, the number of hits (approx. 2,500 articles) shows that an evaluation of all full-text versions would have been logistically impossible with the given means.

Literature assessment

To assess the quality of the included literature, the working group opted for the use of recognised check-lists. The advantage of such tools with broad applicability lies in their standardised structure. Their disadvantage may be an insufficient evaluation of individual study features which may be better recorded with more specific tools.

Literature evaluation

The desire for standardisation and concise representation of actually available details results in a potential loss of some of these details. However, this is counterbalanced by the added value of a systematic review. Independent of methodological or result-related data, it is thus possible to show the important trends in the effects of an intervention.

Conclusion

In children two to 17 years of age, the use of LAIV can lead to a reduction of the number of influenza cases and the associated burden of disease. Additionally, supplementary protection effects may be expected, especially among elderly age groups. However, as for all vaccinations, the achievement of high vaccination coverage rates is crucial in order to reach this target. Currently, no sufficient data is available for the German healthcare setting. Long-term direct and indirect effectiveness and safety should be supported by surveillance programs with a broader use of LAIV.

Since there is no generally accepted model available for the German healthcare setting, statements regarding the cost-effectiveness can be made only with precaution. In this respect, there is additional need to conduct health economic studies concerning the impact of influenza vaccinations for children in Germany. Such studies should be based on dynamic transmission models as these models address indirect protective effects of vaccination correctly.

With regard to ethical, social and legal aspects, there seems to be a need for further analysis of parental/guardian motivation in childhood vaccination decisions and of possible barriers to comprehensive vaccination coverage, particularly in respect of the German population.

This HTA provides an extensive basis for further scientific approaches and pending decisions relating to health policy.

LAIV has recently been licensed as a tetra-/quadrivalent vaccine in Germany. It is intended to ensure protection against both of the Influenza-B strains currently in global circulation (Yamagata and Victoria) and could thus show a greater efficacy compared to placebo. More detailed studies on efficacy, effectiveness and efficiency for a better assessment of this vaccine would be desirable.

Notes

Competing interests

The compilation of the HTA report was supported by a funding of the AstraZeneca GmbH, Wedel to the Herescon GmbH. The authors had the entire formal and substantial control about the HTA report.