In Turkey, the prevalence of diabetes is high but the influenza vaccination coverage rate (VCR) is low (9.1% in 2014), despite vaccination being recommended and reimbursed. This study evaluated the cost-effectiveness of increasing the influenza VCR of adults with type 2 diabetes in Turkey to 20%.
A decision-analytic model was adapted to Turkey using data derived from published sources. Direct medical costs and indirect costs due to productivity loss were included in the societal perspective. The time horizon was set at 1 year to reflect the seasonality of influenza.
Increasing the VCR for adults with type 2 diabetes to 20% is predicted to avert an additional 19,777 influenza cases, 2376 hospitalizations, and 236 deaths. Associated influenza costs avoided were estimated at more than 8.3 million Turkish Lira (TRY), while the cost of vaccination would be more than TRY 8.4 million. The incremental cost-effectiveness ratio was estimated at TRY 64/quality-adjusted life years, which is below the per capita gross domestic product of TRY 21,511 and therefore very cost-effective according to World Health Organization guidelines. Factors most influencing the incremental cost-effectiveness ratio were the excess hospitalization rate, inpatient cost, vaccine effectiveness against hospitalization, and influenza attack rate. Increasing the VCR to >20% was also estimated to be very cost-effective.
Citation: Akın L, Macabéo B, Caliskan Z, Altinel S, Satman I (2016) Cost-Effectiveness of Increasing Influenza Vaccination Coverage in Adults with Type 2 Diabetes in Turkey. PLoS ONE 11(6): e0157657. https://doi.org/10.1371/journal.pone.0157657
Editor: Ray Borrow, Public Health England, UNITED KINGDOM
Received: January 13, 2016; Accepted: June 2, 2016; Published: June 20, 2016
Copyright: © 2016 Akın et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Data underlying the findings described in the manuscript are freely available within the manuscript.
Funding: Financial support for the study and for medical writing was provided by Sanofi Pasteur. The funder participated in study design, data collection and analysis, the decision to publish, and preparation of the manuscript.
Competing interests: BM is an employee of Sanofi Pasteur. SA was employed by Sanofi Pasteur when the study was conducted. IS previously received an unrestricted research grant from Sanofi Pasteur unrelated to this study. This does not alter the authors' adherence to PLOS ONE policies on sharing data and materials. LA and ZC declare no conflicts of interest.
Influenza is a frequent, serious infectious disease that can lead to severe illness, hospitalization, and death . The risk of influenza and its complications, including hospitalization and death, are increased in patients with diabetes . Also, influenza itself aggravates diabetes by increasing the risk of hypoglycemia . This puts increased pressure on the healthcare system and creates a significant public health and economic burden.
In patients with diabetes, influenza vaccination is effective and safe and significantly decreases influenza-related complications and deaths [4–7]. The World Health Organization (WHO) therefore recommends a vaccination coverage rate (VCR) of at least 75% for patients with diabetes . In contrast to some other at-risk populations, influenza vaccine effectiveness is not reduced in patients with diabetes [4, 9–11]. Most health economics studies show that influenza vaccination is cost-effective, especially in at-risk populations , but limited evidence is available for patients with diabetes. Such data are important for encouraging health authorities and physicians to provide influenza vaccination to patients with diabetes and avoiding influenza-associated complications, hospitalizations, and deaths in these patients.
Diabetes is especially common in Turkey. The TURDEP-I cross-sectional survey, conducted in 1997–1998, found a crude prevalence of diabetes of 7.5% (95% CI, 6.0–9.0%) in Turkish adults ≥ 20 years of age . A follow-up survey, TUDEP-II, conducted in 2010, found a prevalence of 16.2% (95% CI, 15.5–21.1%) . Modeling based on the data collected from the TURDEP-I and -II surveys forecasted an increase to 31.5% by 2025 due to increasing rates of obesity and overweight . Although influenza vaccination is recommended for patients with diabetes and other high-risk groups and is entirely reimbursed in Turkey, the VCR for patients with diabetes is reported to be only 9.1% .
In this study, we used a decision-analytic model to evaluate the cost-effectiveness of increasing the influenza VCR for adults with diabetes in Turkey. Although WHO recommendations are a VCR of at least 75% for patients with diabetes, we considered such an increase from the current situation of 9.1% VCR to be unattainable in the immediate future. We therefore examined the cost-effectiveness of an incremental increase, to a projected 20%, as the primary objective. The results of this analysis should add to the very limited health economics data on influenza vaccination in patients with diabetes.
Research Design and Methods
A decision-analytic model  was used to compare costs and outcomes associated with different influenza VCRs for adults with diabetes in Turkey. The model was primarily populated with data obtained from the literature and from the Turkish Statistical Institute and some sensitivity analyses.
Depending on their vaccination status, the model assigns each patient with diabetes specific risks for having uncomplicated influenza (laboratory-confirmed), being admitted to hospital due to major complications of influenza, and dying due to complications of influenza (Fig 1).
(A) Decision tree of patients with diabetes. The projected situation differs from the actual situation only in higher vaccination coverage. Each year, vaccinated and unvaccinated patients with diabetes in both situations have a specific risk of being infected by influenza, which depends on the attack rate and the vaccine effectiveness for those vaccinated. Decision nodes are indicated with squares, and chance nodes are indicated with circles. (B) Resource utilization pathway. GP, general practitioner.
For definition of influenza cases, we used laboratory-confirmed influenza. Influenza cases, hospitalizations, and deaths were modeled independently. Thus, the model was parameterized by the incidence of influenza-associated hospitalizations and influenza-associated mortality and not by the probability of hospitalization conditional on influenza or the case fatality rate. This approach limits the impact of over- or under-estimating the influenza attack rate or the incidence of hospitalizations and mortality.
Healthcare resources to treat influenza and its complications were considered as described previously . Briefly, healthcare resources considered were for outpatient visits and hospitalizations, and the following assumptions were made: patients with major complications had at least one outpatient visit before hospitalization and patients who died had been hospitalized before dying.
The timeframe of the analysis was one year to reflect the seasonality of influenza. However, in calculating life-years (LYs) gained due to avoided mortality, the full span of life expectancy was considered, not only the study year.
The following types of healthcare and productivity loss costs were considered: vaccination cost including the cost of vaccine acquisition and administration; outpatient visit cost for influenza cases; hospitalization cost due to influenza; productivity loss cost due to influenza. All deaths were assumed to occur in patients who had been hospitalized and we did not account for any medical cost specifically associated with death, nor for cost of lost productivity in case of death. The loss of productivity for presenteeism was not taken into account because the actual impact on production was difficult to measure. Also, productivity loss due to early retirement and premature mortality were not included in indirect costs.
The model accounted for two types of negative effects of influenza in terms of quality-adjusted life years (QALYs), namely, QALYs lost due to illness and QALYs lost due to deaths. For the latter, potential years of life lost due to influenza were weighted with utilities to be converted in terms of QALYs. Accordingly, the model required two types of utility values: utility values reflecting quality of life during episodes of influenza, according the presence of complications; and utility norms for the patients with diabetes, by age, applied to LYs.
Potential adverse effects of vaccination were not represented in the model. These are generally local and transient reactions with no associated cost and little impact on quality of life [10, 17, 18].
PubMed was searched for publications relating to influenza in patients with diabetes. Articles were selected if they presented vaccine effectiveness in patients with diabetes, attack rates or complications of influenza in patients with diabetes (e.g. hospitalizations and deaths), VCR, or economic evaluations. Data unavailable through this search were identified by a search via Google and the websites for the WHO and International Diabetes Federation. Data from Turkey were preferred, but when unavailable, data from other countries were used. Data on population size, life expectancy, percentage of population working, and mean daily earning were obtained from the Turkish Statistical Institute, the OECD, or international economic survey data. When data specific to patients with diabetes were not available, data from the general population were used. Further details of the model inputs are described in Table 1.
Population and life expectancy.
The size of Turkish general population aged ≥18 years in 2014 was obtained from the Turkish Statistical Institute , and the population of patients with diabetes in Turkey was estimated from 2010 prevalence data .
Estimates of life expectancy by sex and 5-year interval were for the general Turkish population . Life expectancy for the population of patients with diabetes was based on Canadian data . The life expectancy of the general Turkish population was reduced by the corresponding life expectancy loss to obtain the life expectancy for patients with diabetes in Turkey. In addition, LYs accumulated in future years were discounted at 3%.
Influenza attack rates were obtained from two systematic reviews [23, 24]. The age-specific average annual rate of true influenza for adults and elderly was assumed to be equal to the average of influenza incidence in the placebo arm of clinical trials. A weighted average of the two values was assumed for the adult population. No difference was assumed for attack rates between adults with and without diabetes. For deterministic sensitivity analysis (DSA), the attack rates were set at 1% to 10%. For probabilistic sensitivity analysis (PSA), a beta distribution was used and distribution parameters were derived from the Cochrane reviews [23, 24] (Table 1).
The increased risk of influenza hospitalization in patients with diabetes for the base case analysis was 6-fold and increased risk of death from influenza was 3-fold . The excess rates of influenza hospitalization and death for the general population were adjusted by the risk ratio for patients with diabetes vs. the general population in the US . For DSA, the lower-bound of the increased risk of influenza hospitalization in patients with diabetes was 3-fold based on US data . This increased risk was applied to the excess hospitalization rate in the general European population . For the upper-bound, the increase in risk was 15.5-fold based on US data , which was applied to the base case excess hospitalization rate . For the excess death rate, the lower-bound in increased risk was 2-fold based on US data  and the upper-bound was 8.3-fold based on Canadian data . For PSA, a beta distribution was used, and distribution parameters were derived from US data  adjusted to patients with diabetes using US values for increased risk  (Table 1).
Estimates of utilities for patients with diabetes without influenza are required to calculate QALYs. Utility norms in the population of patients with diabetes by age group for the base case analysis were based on Canadian data . In the DSA, ±20% was tested. Utility norms were not included for the PSA.
Utility values for influenza were from Rothberg et al. . In the study, 15 randomly selected working-age adults and healthcare workers with a history of influenza were asked to describe their health status during illness using the Health Utilities Index (HUI-3) instrument. The resulting utilities were 0.25 for uncomplicated influenza (illness) and 0.2 for complicated influenza (hospitalization) . Utility values for uncomplicated influenza from previous models were generally based on expert opinion and were between 0.52 and 0.64 [33–36]. We therefore conducted a sensitivity analysis with an upper-bound utility of 0.64 and a lower-bound of −20% for uncomplicated influenza (illness). For complicated influenza, we performed sensitivity analysis using ±20% and the upper- and lower-bounds.
Resource utilization and costs.
The model took into account that a substantial proportion of patients with influenza do not seek medical attention. The proportion of outpatient visits was based on probability to seek medical attention given influenza in high-risk people in US (62.5% for 18–64 y and 82% for >65 y) . For uncomplicated influenza, the duration of influenza symptoms was 3.53 days . The duration of influenza-associated hospitalization in patients with diabetes in Turkey was estimated as 7.2 days based on data from a hospital-based surveillance network . For DSA of the duration of uncomplicated influenza, −20% was used for the lower bound and 11.5 days for the upper bound . For PSA, a lognormal distribution was used. For DSA of the duration of complicated influenza, the lower-bound was considered to be equal to the duration of uncomplicated influenza and the upper-bound to be 12 days . For PSA, a lognormal distribution was used.
Based on WHO-CHOICE estimates , the cost per outpatient visit was estimated at 17.81 Turkish Lira (TRY) in 2008 and inflated to 2014 values (0.43 per US dollar) using the Consumer Price Index . For DSA of outpatient cost, ±20% was used for the lower and upper bound. For PSA, a lognormal distribution was used.
The average cost for patients with diabetes hospitalized with severe influenza disease in Turkey was estimated to be TRY 2079 in 2014 based on data from hospital-based surveillance network . For DSA of inpatient cost, the lower bound was based on Turkish data  and the upper-bound was based on data from WHO-CHOICE . For PSA, a lognormal distribution was used.
Although there is no any gold standard to measure productivity loss due to an illness, the sources of lost productivity are assumed to include absence from paid work (absenteeism), decreased productivity at work (presentism), early retirement due to illness, early mortality for employee, and unpaid production. In this study, the only source of lost productivity was assumed to be absenteeism. Absenteeism costs due to influenza were calculated using the human capital approach. In this approach, the value of lost productivity is calculated as the product of the number of lost workdays and the cost of one workday from an employer’s perspective. The number of working days lost was estimated 2.7 days for uncomplicated influenza using US data  and a systematic review  and 7.2 days for complicated influenza using Turkish data for hospitalization . For workdays lost due to uncomplicated influenza, the upper and lower bounds of the DSA (1.5 and 4.9 days) were estimated from the systematic review  and a lognormal distribution was used for PSA. Mean daily earnings were estimated from the Structure of Earnings Survey 2010 from the Turkish Statistical Institute  and inflated to 2014 using the Consumer Price Index.
In Turkey, patients with diabetes are fully reimbursed for outpatient and inpatient costs.
The VCR used for the current situation was 9.1% for patients with diabetes in Turkey . For the main analysis, the new VCR tested was 20%. VCRs of 30%, 40%, 50%, and 75% were also tested.
The vaccine effectiveness against laboratory-confirmed influenza was assumed to be similar for people with or without diabetes based on several previous studies showing that humoral and cell mediated immune responses are comparable in patients with or without diabetes and that immune response to vaccination is appropriate in patients with diabetes [4, 9–11]. Vaccine effectiveness against laboratory-confirmed influenza in adults aged 18 years and older was 55% [44, 45]. A range of 40% to 60% was used for the DSA. Vaccine effectiveness against hospitalization (54%, 26–71% in DSA) and death (58%, 13–80% in DSA) was based on Dutch data . For PSA, a lognormal distribution was used for vaccine effectiveness against laboratory-confirmed influenza, hospitalization, and death.
The public reimbursed vaccine price was TRY 9.42 in 2014, and patients with diabetes are fully reimbursed for the influenza vaccination. There is no administration cost in public in Turkey because it is routine practice for primary health care services.
The model was developed in Excel 2003 (Microsoft Corporation, Redmond, WA). Design of the model was similar to a previous report  and included the following outcomes: influenza cases, hospitalizations, and deaths; QALYs; and life years gained (i.e., improvement in life expectancy). Based on recommendations of the US Panel on Cost-Effectiveness in Health and Medicine , a 3% discount rate was used for LYs and QALYs accumulated over future years. Sensitivity analyses were performed using discount rates of 0% and 5%.
The cost of productivity loss due to influenza, valued using the human capital approach, was included for the analysis from a societal perspective, whereas only the direct medical costs, which are fully reimbursed for patients with diabetes, were included for the analysis from a public payer perspective. Discounting costs was not necessary because all costs were incurred within a single year. Costs were calculated in TRY.
Cost-effectiveness ratios including incremental cost per QALY gained were calculated.
DSA was performed to identify the main sources of uncertainty in the model and their impact. Sensitivity of the base case incremental cost-effectiveness ratio to input parameters were explored by varying key parameters within ranges reflecting possible parameter values (Table 1).
To evaluate the impact of simultaneous variation of input parameters on the model outcomes, PSA using Monte Carlo simulation was performed. Distribution parameters were defined according to the assumed distribution of each key input parameter (Table 1). A total of 1000 iterations were performed, and for each combination of parameters the difference of costs and QALY gained were displayed in a cost-effectiveness plane.
Because Turkish guidelines do not establish a threshold below which a new product is considered cost-effective, WHO guidelines were used to determine the cost-effectiveness ratio. Cost-effectiveness ratios were compared with the yearly gross domestic product (GDP) per capita in Turkey, which, based on purchasing-power-parity per capita  and the purchasing-power-parity conversion factor , was estimated at TRY 21,511 for 2014. As defined by the WHO guidelines, a vaccination strategy was considered very cost-effective if its incremental cost-effectiveness ratio (ICER) is below the yearly GDP per capita and cost-effective if the ICER is between 1 and 3 yearly GDP per capita .
Base-case analysis from a societal perspective
Increasing the influenza VCR from the current situation of 9.1% to a projected 20% would avoid an estimated additional 19,777 cases, 2376 hospitalizations, and 236 deaths due to influenza in patients with diabetes (Table 2). This increase in VCR would gain an additional 2743 QALYs and 3342 LYs. An estimated TRY 8,314,636 of influenza costs would be avoided, while vaccination costs associated with the increase in VCR would be TRY 8,489,233, resulting in an estimated cost difference of TRY 174,597. Based on these values, the ICER is estimated to be TRY 64/QALY from a societal perspective. This is below the per capita GDP in Turkey (TRY 21,511), so changing from the current situation to the new situation of 20% coverage is considered very cost-effective based on the WHO definition .
Sensitivity analysis from a societal perspective: DSA and PSA
One-way DSA was performed using the lower- and upper-bound of parameters or, for cases where the upper- and lower-bounds were not set, a range of ±20%. DSA shows that the ICER was most sensitive to the excess hospitalization rate, with an ICER of TRY 1699/QALY using the lower-bound and a cost-saving result using the upper-bound (TRY −3025/QALY). The other parameters with the greatest influence on the ICER were inpatient cost, vaccine effectiveness against hospitalization, attack rate, and vaccine effectiveness against influenza cases and against death (Fig 2 and S1 Table). The maximum ICER was below the per capita GDP threshold in Turkey, so all scenarios were considered very cost-effective.
One-way deterministic sensitivity analysis was performed using the lower-bound (in gray) and upper-bound (in black) of parameters or, for cases where the upper- and lower-bounds were not set, a range of ± 20%. Abbreviations: TRY, Turkish Lira; QALY, quality-adjusted life years.
PSA using Monte Carlo simulation was performed to evaluate the impact of simultaneous variation of input parameters on the model outcomes. Outputs of 1000 simulations are presented in the cost-effectiveness plane (Fig 3). The scatter represents the different outcomes of the 1000 simulations from probabilistic sensitivity analysis based on a range of different input values. All of the simulations were below the willingness-to-pay threshold of 1 per capita GDP/QALY, meaning that there is a probability of 100% that, from the societal perspective, the new situation (20% VCR) would be very cost-effective compared with the current vaccination situation (9.1% VCR). Of 1000 simulations, 10 (1.0%) were very cost-effective (ICER <1 per capita GDP) and 990 (99.0%) were cost-neutral (ICER = 0) or cost-saving (ICER<0).
Probabilistic sensitivity analysis using Monte Carlo simulation was performed to evaluate the impact of simultaneous variation of input parameters on the model outcomes. Outputs of 1000 simulations are presented in the cost-effectiveness plane. The black line represents a willingness-to-pay threshold of 21,511 Turkish Lira (TRY) (per capita gross domestic product) per quality-adjusted life year (QALY). Dots above the X-axis are very cost-effective (incremental cost-effectiveness ratio < per capita gross domestic product), those on the X-axis are cost-neutral (incremental cost-effectiveness ratio = 0), and those below the X-axis are cost-saving (incremental costs < TRY 0 are negative while incremental QALYs remain positive).
Cost-effectiveness of the base case from a public payer perspective
The ICER associated with moving from the current VCR of 9.1% to a projected VCR of 20% was estimated to be TRY 1158/QALY from a public payer perspective. This is also considered very cost-effective according to the WHO definition.
Secondary analysis–impact of increasing VCRs beyond 20% and up to the recommended coverage of 75% for patients with diabetes
The WHO  and the European council  recommend a VCR of 75% for patients with diabetes. We therefore examined additional projected situations in which the VCR was increased to 30–75%. By increasing from the current situation of 9.1% VCR to 30–75% VCR, an estimated 37,920–119,567 additional cases, 4556–14,367 hospitalizations, and 452–1426 deaths would be avoided each year (Table 2). QALYs gained would be 5258–16,580 and LY gained would be 6409–20,208. The projected situations of 30–75% VCR would translate into vaccination cost increases of TRY 16.3 to 51.3 million but would avoid TRY 15.9 to 50.2 million of influenza-related costs, resulting in an overall cost from a societal perspective of TRY 334,778–1,055,590. The resulting ICER remained at TRY 64/QALY for all projected VCRs. Thus, increasing the influenza VCR from the current situation of 9.1% to 30–75% would still be considered very cost-effective according to WHO guidelines.
In the current study, we examined the cost-effectiveness of increasing influenza vaccination coverage in Turkish adults with diabetes. Using a decision-analytic model, we showed that increasing the influenza VCR in Turkish adults with diabetes from the current 9.1% to 20% would be very cost-effective. Higher VCRs, between 30% and the WHO- and European-recommended rate of 75% for at-risk populations [50, 51], would also be very cost-effective. Such increases in VCR should be attainable because a one-day training of specialist physicians in Turkey increased the VCR in adults with diabetes more than two fold .
Two important strengths of this study are that we estimated the cost-effectiveness of increases in influenza VCRs in patients with diabetes and that we examined this in a developing country. Relatively few studies have been performed to assess cost-effectiveness of influenza vaccination in patients with underlying conditions, especially those with diabetes. Influenza vaccination has been shown to be cost-effective in patients with other underlying conditions, such as chronic obstructive pulmonary disease , coronary heart disease , cancer , hemodialysis , and human immunodeficiency virus infection . Although the literature focusing on patients with diabetes is limited, influenza vaccination has been shown to be cost-effective in the US . Using a Markov model, the US study estimated that for patients with diabetes ≥25 years of age, the ICER for influenza vaccination was 6083 US dollars/QALY, which is considered very cost-effective according to WHO guidelines.
One of the few studies to look at the cost-effectiveness of influenza vaccination in patients with diabetes in developing countries was published by Narayan et al. . Based on a comprehensive review of data published up to 2003 and assuming that QALYs would be the same as in developed countries but assuming different costs, they estimated a cost per QALY for influenza vaccination of 310 US dollars for elderly adults with type 2 diabetes in the Middle East and North Africa. Thus, our conclusions concur with and extend the limited information on the cost-effectiveness of influenza vaccination in patients with diabetes in developing countries.
Another important strength of this study is that we used a model that allowed us to examine influenza-associated hospitalizations and deaths independently. Thus, over- or under-estimation of the influenza attack rate would not necessarily imply over- or under-estimation of the incidence of hospitalization and mortality. The model also allowed for different effectiveness rates against influenza cases, influenza-associated hospitalizations, and deaths.
A potential limitation of this study is that we had to ignore some factors, although the factors that we ignored should have little influence on the final conclusions. We assumed that deaths occurred in hospitalized cases; that adverse effects of vaccination would be negligible; and that herd immunity would have little impact at the VCRs assessed, especially at the low VCR of 20% in the base case. In addition, we did not take into account costs and probabilities of clinical events specific to patients with diabetes. We also ignored the medical cost and the cost of lost productivity in cases of death, as well as productivity loss due to presenteeism and the cost of prescription and over-the-counter drugs. Because we did not include these costs, our conclusions are based on conservative cost estimates. Furthermore, influenza vaccination was also very cost-effective even when productivity loss costs due to influenza were excluded. Finally, although several studies have shown that patients with diabetes are at increased risk of influenza and other lower respiratory tract infections , due to a lack of robust evidence, we had to assume no difference in risk between adults with and without diabetes. This could have led to an underestimate of the attack rate, the rate of complications, or both. Despite these potential effects, DSA and PSA confirmed that influenza vaccination was very cost-effective in this population across a wide range of attack and vaccine effectiveness rates.
Another limitation of this study is that, as in other cost-effectiveness studies for developing countries , data were often lacking, especially for patients with diabetes, and had to be extrapolated from developed countries. In particular, to estimate the impact of diabetes on population health, we used a Canadian study that estimated life expectancy and health-adjusted life expectancy for patients with and without diabetes . In addition, the excess hospitalization and death rates due to influenza in the general population was based on US data , but in sensitivity analyses, we tested the lower-bound using the excess hospitalization and death rates in Europe . Although the excess hospitalization rate was the strongest factor influencing cost-effectiveness, influenza vaccination remained very cost-effective across a wide range of excess risk values. Utility values  and productivity loss estimates [26, 42] were also from developed countries, but DSA showed that variation of these values had a negligible effect on the cost-effectiveness. Finally, based on US data, we assumed a conservative 3% discount rate for LYs accumulated in future years to avoid over-estimating life years saved by influenza vaccination.
This study showed that, in Turkey, influenza vaccination of patients with diabetes is very cost-effective. Considering the overall expenditure for diabetes management [58, 62, 63], the budget impact of influenza vaccination for patients with diabetes is a reasonable additional investment. The public health benefits of influenza vaccination, however, will only be realized if immunization programs and targeted interventions are implemented by local authorities. To reduce the burden of influenza in patients with diabetes, an integrated care approach is needed and all stakeholders must cooperate to increase the VCRs to WHO-recommended levels.
Conceived and designed the experiments: BM IS ZC. Performed the experiments: BM. Analyzed the data: BM IS LA ZC SA. Contributed reagents/materials/analysis tools: BM ZC. Wrote the paper: BM IS LA ZC SA.
- 1. Monto AS. Seasonal influenza and vaccination coverage. Vaccine. 2010;28 Suppl 4:D33–44. Epub 2010/08/18. pmid:20713259.
- 2. Mertz D, Kim TH, Johnstone J, Lam PP, Science M, Kuster SP, et al. Populations at risk for severe or complicated influenza illness: systematic review and meta-analysis. BMJ. 2013;347:f5061. Epub 2013/08/27. pmid:23974637; PubMed Central PMCID: PMC3805492.
- 3. Shafiee G, Mohajeri-Tehrani M, Pajouhi M, Larijani B. The importance of hypoglycemia in diabetic patients. Journal of diabetes and metabolic disorders. 2012;11(1):17. Epub 2012/01/01. pmid:23497433; PubMed Central PMCID: PMC3598174.
- 4. Colquhoun AJ, Nicholson KG, Botha JL, Raymond NT. Effectiveness of influenza vaccine in reducing hospital admissions in people with diabetes. Epidemiology and infection. 1997;119(3):335–41. Epub 1998/01/24. pmid:9440437; PubMed Central PMCID: PMC2809006.
- 5. Looijmans-Van den Akker I, Verheij TJ, Buskens E, Nichol KL, Rutten GE, Hak E. Clinical effectiveness of first and repeat influenza vaccination in adult and elderly diabetic patients. Diabetes care. 2006;29(8):1771–6. Epub 2006/07/29. pmid:16873778.
- 6. Nichol KL, Nordin J, Mullooly J, Lask R, Fillbrandt K, Iwane M. Influenza vaccination and reduction in hospitalizations for cardiac disease and stroke among the elderly. The New England journal of medicine. 2003;348(14):1322–32. Epub 2003/04/04. pmid:12672859.
- 7. Schade CP, McCombs MA. Influenza immunization and mortality among diabetic Medicare beneficiaries in West Virginia. The West Virginia medical journal. 2000;96(3):444–8. Epub 2003/11/19. pmid:14619136.
- 8. World Health Organization. Prevention and control of influenza pandemics and annual epidemics Geneva: World Health Organization; 2003.
- 9. Smith SA, Poland GA. Immunization and the prevention of influenza and pneumococcal disease in people with diabetes. Diabetes care. 2003;26 Suppl 1:S126–8. Epub 2002/12/28. pmid:12502639.
- 10. Feery BJ, Hartman LJ, Hampson AW, Proietto J. Influenza immunization in adults with diabetes mellitus. Diabetes care. 1983;6(5):475–8. Epub 1983/09/01. pmid:6400708.
- 11. Pozzilli P, Gale EA, Visalli N, Baroni M, Crovari P, Frighi V, et al. The immune response to influenza vaccination in diabetic patients. Diabetologia. 1986;29(12):850–4. Epub 1986/12/01. pmid:3569690.
- 12. Peasah SK, Azziz-Baumgartner E, Breese J, Meltzer MI, Widdowson MA. Influenza cost and cost-effectiveness studies globally—a review. Vaccine. 2013;31(46):5339–48. Epub 2013/09/24. pmid:24055351.
- 13. Satman I, Yilmaz T, Sengul A, Salman S, Salman F, Uygur S, et al. Population-based study of diabetes and risk characteristics in Turkey: results of the turkish diabetes epidemiology study (TURDEP). Diabetes care. 2002;25(9):1551–6. Epub 2002/08/28. pmid:12196426.
- 14. Sozmen K, Unal B, Capewell S, Critchley J, O'Flaherty M. Estimating diabetes prevalence in Turkey in 2025 with and without possible interventions to reduce obesity and smoking prevalence, using a modelling approach. International journal of public health. 2015;60 Suppl 1:S13–21. Epub 2014/12/05. pmid:25471076.
- 15. Ciblak MA. Influenza vaccination in Turkey: prevalence of risk groups, current vaccination status, factors influencing vaccine uptake and steps taken to increase vaccination rate. Vaccine. 2013;31(3):518–23. Epub 2012/11/24. pmid:23174194.
- 16. Aballea S, Chancellor J, Martin M, Wutzler P, Carrat F, Gasparini R, et al. The cost-effectiveness of influenza vaccination for people aged 50 to 64 years: an international model. Value in health: the journal of the International Society for Pharmacoeconomics and Outcomes Research. 2007;10(2):98–116. Epub 2007/03/30. pmid:17391419.
- 17. el-Madhun AS, Cox RJ, Seime A, Sovik O, Haaheim LR. Systemic and local immune responses after parenteral influenza vaccination in juvenile diabetic patients and healthy controls: results from a pilot study. Vaccine. 1998;16(2–3):156–60. Epub 1998/06/02. pmid:9607024.
- 18. Turner D, Wailoo A, Nicholson K, Cooper N, Sutton A, Abrams K. Systematic review and economic decision modelling for the prevention and treatment of influenza A and B. Health Technol Assess. 2003;7(35):iii-iv, xi-xiii, 1–170. Epub 2003/11/12. pmid:14609480.
- 19. Turkish Statistical Institute. Population of Province/District Centers and Towns/Villages by Age Group and Sexm 2007–2013. Available from: http://www.turkstat.gov.tr/PreIstatistikTablo.do?istab_id=1587.
- 20. Satman I, Omer B, Tutuncu Y, Kalaca S, Gedik S, Dinccag N, et al. Twelve-year trends in the prevalence and risk factors of diabetes and prediabetes in Turkish adults. European journal of epidemiology. 2013;28(2):169–80. Epub 2013/02/15. pmid:23407904; PubMed Central PMCID: PMC3604592.
- 21. Expectancy WL. Health Profile: Turkey.
- 22. Loukine L, Waters C, Choi BC, Ellison J. Impact of diabetes mellitus on life expectancy and health-adjusted life expectancy in Canada. Population health metrics. 2012;10(1):7. Epub 2012/04/26. pmid:22531113; PubMed Central PMCID: PMC3787852.
- 23. Jefferson T, Di Pietrantonj C, Al-Ansary LA, Ferroni E, Thorning S, Thomas RE. Vaccines for preventing influenza in the elderly. The Cochrane database of systematic reviews. 2010;(2):CD004876. Epub 2010/02/19. pmid:20166072.
- 24. Jefferson T, Di Pietrantonj C, Rivetti A, Bawazeer GA, Al-Ansary LA, Ferroni E. Vaccines for preventing influenza in healthy adults. The Cochrane database of systematic reviews. 2010;(7):CD001269. Epub 2010/07/09. pmid:20614424.
- 25. US Centers for Disease Control and Prevention. If You Have Diabetes, A Flu Shot Could Save Your Life. Atlanta: US CDC; 2015.
- 26. Molinari NA, Ortega-Sanchez IR, Messonnier ML, Thompson WW, Wortley PM, Weintraub E, et al. The annual impact of seasonal influenza in the US: measuring disease burden and costs. Vaccine. 2007;25(27):5086–96. Epub 2007/06/05. pmid:17544181.
- 27. Suryaprasad A, Redd JT, Hancock K, Branch A, Steward-Clark E, Katz JM, et al. Severe acute respiratory infections caused by 2009 pandemic influenza A (H1N1) among American Indians—southwestern United States, May 1-July 21, 2009. Influenza and other respiratory viruses. 2013;7(6):1361–9. Epub 2013/06/01. pmid:23721100.
- 28. Preaud E, Durand L, Macabeo B, Farkas N, Sloesen B, Palache A, et al. Annual public health and economic benefits of seasonal influenza vaccination: a European estimate. BMC public health. 2014;14:813. Epub 2014/08/12. pmid:25103091; PubMed Central PMCID: PMC4141103.
- 29. Mullooly JP, Bridges CB, Thompson WW, Chen J, Weintraub E, Jackson LA, et al. Influenza- and RSV-associated hospitalizations among adults. Vaccine. 2007;25(5):846–55. Epub 2006/11/01. pmid:17074423.
- 30. Schaffner W, Rehm SJ, Elasy TA. Influenza Vaccination: An Unmet Need in Patients With Diabetes. Clinical Diabetes. 2007;25(4):145–9
- 31. Schanzer DL, Langley JM, Tam TW. Co-morbidities associated with influenza-attributed mortality, 1994–2000, Canada. Vaccine. 2008;26(36):4697–703. Epub 2008/07/16. pmid:18620016.
- 32. Rothberg MB, He S, Rose DN. Management of influenza symptoms in healthy adults. Journal of general internal medicine. 2003;18(10):808–15. Epub 2003/10/03. pmid:14521643; PubMed Central PMCID: PMC1494927.
- 33. O'Brien BJ, Goeree R, Blackhouse G, Smieja M, Loeb M. Oseltamivir for treatment of influenza in healthy adults: pooled trial evidence and cost-effectiveness model for Canada. Value in health: the journal of the International Society for Pharmacoeconomics and Outcomes Research. 2003;6(2):116–25. Epub 2003/03/19. pmid:12641862.
- 34. Burls A, Clark W, Stewart T, Preston C, Bryan S, Jefferson T, et al. Zanamivir for the treatment of influenza in adults: a systematic review and economic evaluation. Health Technol Assess. 2002;6(9):1–87. Epub 2002/06/18. pmid:12065066.
- 35. Mauskopf JA, Cates SC, Griffin AD, Neighbors DM, Lamb SC, Rutherford C. Cost effectiveness of zanamivir for the treatment of influenza in a high risk population in Australia. PharmacoEconomics. 2000;17(6):611–20. Epub 2000/09/08. pmid:10977398.
- 36. Brady B, McAuley L, Shukla VK. Economic Evaluation of Zanamivir for the Treatment of Influenza. Ottawa: Canadian Coordinating Office for Health Technology Assessment, 2001 Contract No.: Technology report no 13.
- 37. Badur S, Ciblak MA, Tanrıöver MD, Özışık L, Şensoy AE, Gencer S, et al. Türkiye’de influenza nedeniyle hastaneye yatırılan diyabetli erişkin hastaların neden olduğu tedavi giderleri. Global hastane bazlı influenza sürveyansı; Antalya, Turkey2015.
- 38. Cooper NJ, Sutton AJ, Abrams KR, Wailoo A, Turner D, Nicholson KG. Effectiveness of neuraminidase inhibitors in treatment and prevention of influenza A and B: systematic review and meta-analyses of randomised controlled trials. BMJ. 2003;326(7401):1235. Epub 2003/06/07. pmid:12791735; PubMed Central PMCID: PMC161558.
- 39. WHO-CHOICE. Health service delivery costs 2015. Available from: http://www.who.int/choice/cost-effectiveness/inputs/health_service/en/.
- 40. OECD. Consumer Prices (MEI). Available from: http://stats.oecd.org/index.aspx?querytype=view&queryname=221.
- 41. Yoldascan E, Kurtaran B, Koyuncu M, Koyuncu E. Modeling the economic impact of pandemic influenza: a case study in Turkey. Journal of medical systems. 2010;34(2):139–45. Epub 2010/05/04. pmid:20433052.
- 42. Keech M, Beardsworth P. The impact of influenza on working days lost: a review of the literature. PharmacoEconomics. 2008;26(11):911–24. Epub 2008/10/15. pmid:18850761.
- 43. Turkish Statistical Institute. Average working hours and monthly average gross wage by sex and age group. The results of Structure of Earnings Survey. Available from: http://www.turkstat.gov.tr/PreTablo.do?alt_id=1008.
- 44. DiazGranados CA, Denis M, Plotkin S. Seasonal influenza vaccine efficacy and its determinants in children and non-elderly adults: a systematic review with meta-analyses of controlled trials. Vaccine. 2012;31(1):49–57. Epub 2012/11/13. pmid:23142300.
- 45. Beyer WE, McElhaney J, Smith DJ, Monto AS, Nguyen-Van-Tam JS, Osterhaus AD. Cochrane re-arranged: support for policies to vaccinate elderly people against influenza. Vaccine. 2013;31(50):6030–3. Epub 2013/10/08. pmid:24095882.
- 46. Russell LB, Gold MR, Siegel JE, Daniels N, Weinstein MC. The role of cost-effectiveness analysis in health and medicine. Panel on Cost-Effectiveness in Health and Medicine. Jama. 1996;276(14):1172–7. Epub 1996/10/09. pmid:8827972.
- 47. International Monetary Fund. Report for Selected Countries and Subjects
- 48. The World Bank. PPP conversion factor, GDP (LCU per international $) Available from: http://data.worldbank.org/indicator/PA.NUS.PPP.
- 49. World Health Organization. Making choices in health: WHO Guide to Cost-Effectiveness Analysis. Geneva: World Health Organization; 2003.
- 50. SAGE Working Group. Background Paper on Influenza Vaccines and Immunization. Geneva: World Health Organization, 2012.
- 51. Council of the European Union. COUNCIL RECOMMENDATION of 22 December 2009 on seasonal influenza vaccination. Official journal of the European Union. 2009;L 348:71–2.
- 52. Satman I, Akalin S, Cakir B, Altinel S. The effect of physicians' awareness on influenza and pneumococcal vaccination rates and correlates of vaccination in patients with diabetes in Turkey: an epidemiological Study "diaVAX". Human vaccines & immunotherapeutics. 2013;9(12):2618–26. Epub 2013/07/28. pmid:23887188; PubMed Central PMCID: PMC4162054.
- 53. Wongsurakiat P, Lertakyamanee J, Maranetra KN, Jongriratanakul S, Sangkaew S. Economic evaluation of influenza vaccination in Thai chronic obstructive pulmonary disease patients. Journal of the Medical Association of Thailand = Chotmaihet thangphaet. 2003;86(6):497–508. Epub 2003/08/20. pmid:12924797.
- 54. Choosakulchart P. Cost-utility evaluation of influenza vaccination in patients with existing coronary heart diseases in Thailand. Asian Biomedicine. 2013;7(3):425–35.
- 55. Avritscher EB, Cooksley CD, Geraci JM, Bekele BN, Cantor SB, Rolston KV, et al. Cost-effectiveness of influenza vaccination in working-age cancer patients. Cancer. 2007;109(11):2357–64. Epub 2007/04/26. pmid:17457827.
- 56. Lee BY, Stalter RM, Bacon KM, Tai JH, Bailey RR, Zimmer SM, et al. Cost-effectiveness of adjuvanted versus nonadjuvanted influenza vaccine in adult hemodialysis patients. American journal of kidney diseases: the official journal of the National Kidney Foundation. 2011;57(5):724–32. Epub 2011/03/15. pmid:21396760; PubMed Central PMCID: PMC3085888.
- 57. Nosyk B, Sharif B, Sun H, Cooper C, Anis AH. The cost-effectiveness and value of information of three influenza vaccination dosing strategies for individuals with human immunodeficiency virus. PloS one. 2011;6(12):e27059. Epub 2011/12/14. pmid:22162988; PubMed Central PMCID: PMC3232195.
- 58. Sorensen S, Gibbons C, Engelgau M, Narayan KMV, Hoerger T, Valdez R, et al., editors. Assessment of the cost-effectiveness of influenza vaccination for type 2 diabetes. Abstract 46-OR. American Diabetes Association Annual Meeting; 2003 13–17 June, 2003; New Orleans.
- 59. Narayan KMV, Zhang P, Kanaya AM, Williams DE, Engelgau MM, Imperatore G, et al. Chapter 30 Diabetes: The Pandemic and Potential Solutions. Disease Control Priorities in Developing Countries. 2nd edition ed. Washington (DC): World Bank; 2006.
- 60. Muller LM, Gorter KJ, Hak E, Goudzwaard WL, Schellevis FG, Hoepelman AI, et al. Increased risk of common infections in patients with type 1 and type 2 diabetes mellitus. Clinical infectious diseases: an official publication of the Infectious Diseases Society of America. 2005;41(3):281–8. Epub 2005/07/12. pmid:16007521.
- 61. Narayan KMV, Zhang P, Kanaya AM, Williams DE, Engelgau MM, Imperatore G, et al. Diabetes: The Pandemic and Potential Solutions In: Jamison DT, Breman GJ, Measham AR, Alleyne G, Claeson M, Evans DB, et al., editors. Disease Control Priorities in Developing Countries. 2nd edition ed. Washington (DC): World Bank; 2006.
- 62. International Diabetes Federation. IDF Diabetes Atlas. Brussels: International Diabetes Federation, 2013.
- 63. [No author listed]. The cost of diabetes. Diabetes UK, 2014.