Influenza is a highly contagious respiratory disease caused by influenza viruses. Some 5% - 10% of adults and 20% - 30% of children worldwide will contract seasonal influenza annually [1] . Children and adults with underlying chronic medical conditions are particularly vulnerable to complications of the infection [2] . Other groups at increased risk of severe disease and complications from influenza include pregnant women, very young child- ren (aged 6 to 23 months) and older adults (commonly persons aged >65 years). Globally, seasonal influenza results in about 5 million severe cases and 500,000 annual deaths each year [1] .

Healthcare workers and nursing home staff constitute a special group at an elevated risk of both contracting and transmitting influenza to already vulnerable patients. Nosocomial infections can compromise quality of care and patient safety and result in prolonged hospital stays, microbial resistance, exacerbations of existing conditions and even deaths [3] . Furthermore, absenteeism from work and resulting working days lost can impose significant economic costs on healthcare systems [4] .

The primary and most effective way to reduce the burden of the disease is seasonal vaccinations. Since the virus mutates from year to year, annual vaccinations are paramount. Maintaining high vaccination coverage is particularly important in patients with chronic non-communicable diseases, or NCDs. NCDs refer to a wide range of diseases that are chronic and progress slowly. Cardiovascular Diseases (CVDs), Cancer, Diabetes and Chronic Obstructive Pulmonary Disease (COPD) comprise the four main groups of NCDs, due to their shared risk factors (tobacco, alcohol, diet and physical activity) and their contribution to the global burden of disease. These four groups of diseases will be implied throughout the paper when NCDs are mentioned.

In 2003, the World Health Assembly recommended that countries should target people at high risk, including older adults and patients with underlying medial conditions, with the goal of attaining influenza vaccination coverage of at least 75% by 2010 [5] .

It has been more than a decade since member states committed to this recommendation, but the challenge has been to translate the recommendations into action. Despite the public awareness and policy efforts made at the global and national levels, influenza vaccination coverage, particularly in high-risk populations, is universally low, and the need to make vaccines available to all those who need them is critical to prevent influenza outbreaks.

Evidence suggests that influenza control efforts focused on high-risk groups in terms of exposure and contagiousness are advantageous for the reduction of influenza burden in comparison to interventions targeting the general populations. More recent recommendations emphasize prioritizing vaccination for individuals at high risk for complications due to influenza virus infection [6] .

Our background research indicates that there is abundant information on the primary prevention and control of major NCDs, but little attention has been paid to the interplay of communicable and non-communicable diseases and the targeted interventions of mutual benefit. There is sufficient data to demonstrate that, apart from the fact that some NCDs have an infectious etiology (e.g. gastric cancer, hepatocellular cancer and cervical cancer), the high burden of infectious diseases and associated chronic inflammation exacerbates risks for other NCDs as well [7] .

Bearing this in mind, we investigated the link between influenza and NCDs and the potential of vaccinations against seasonal influenza as secondary prevention for NCDs. Based on the available evidence in the literature, we believe that persons living with NCDs while contracting the influenza virus may face double vulnerabilities if they are a child, elderly, pregnant, or working in health care. Hence, novel approaches are needed to design strategies that explicitly address this increased vulnerability to the influenza virus.

Therefore, the purpose of this paper is threefold: 1) to alert the global health community to the relationship between seasonal influenza and NCDs; 2) to underscore the benefits of influenza vaccination as secondary prevention for NCDs and 3) to propose strategies for increasing vaccination coverage in this special high-risk group.

NCDs are the most important cause of premature mortality globally, accounting for 36 million (63%) of the 57 million annual deaths [8] , and will have an estimated associated cost of over $30 trillion over the next 20 years [9] . Nearly 80% of NCD mortality takes place in low- and middle-income countries, where health systems are ill prepared to deal with the NCD burden effectively [8] .

There is sufficient epidemiological evidence confirming that influenza is associated with higher rates of complications, hospitalizations and even deaths in individuals living with NCDs versus the general population. This was particularly notable during the 2009-2010 H1N1 pandemic [10] . In non-pandemic years, 80% of persons hospitalized for influenza in the USA had one or more underlying medical conditions [11] .

In terms of complications and hospitalizations, the probability of hospitalization for influenza is three times higher in diabetics than in the general population [12] . In persons with cardiovascular disease, systemic respiratory infections, which are frequently caused by influenza viruses, increase the risk of stroke and heart attacks three- and five-fold respectively, in the three days following the onset of infection [13] .

The impact of influenza infection on mortality among high-risk groups is even more pronounced. In the UK, epidemiological surveillance data from 2010 to 2011 indicated that patients in a risk group due to chronic NCDs had a 10-fold greater risk of mortality due to influenza compared to patients who were not in an at-risk category [14] . In general, peak periods of mortality among NCD patients coincide with peaks of pneumonia and seasonal influenza [15] . As an example, the risk of dying from acute myocardial infarction and chronic ischaemic heart disease is 1.3 times greater during influenza epidemic weeks [16] . Case fatality rates from Influenza A can be over 30% in persons with COPD compared to 0.1% or less in the healthy population [17] . Mortality rates in NCD patients range from ten to 377 per 100,000 influenza cases, depending on the number of high-risk conditions [18] .

The highest rates of influenza mortality are noted among persons aged 65 and older. For example, in the United States influenza infection was responsible for 132.5 per 100,000 person-years for all cause deaths, 98.3 for underlying respiratory and circulatory deaths and 22.1 for underlying pneumonia and influenza deaths [19] . If two comorbid conditions are present (old persons and high-risk), influenza-related death rates are 100 times greater than in healthy adults [20] .

There is sufficient evidence to demonstrate that vaccination against influenza is one of the most cost-effective public health interventions. According to WHO, vaccination can reduce influenza-related morbidity by 60% and mortality by up to 80%. In addition, the indirect benefits from vaccinations may include savings in terms of reduced related healthcare costs due to reduced disease burden [1] .

Vaccination against influenza is particularly beneficial for persons living with NCDs with regard to reduced mortality, hospitalizations and complications. When the impact of vaccination on NCDs was measured in over 35,000 older adults, mortality from stroke, diabetes, COPD, and heart disease was lowered by 65%, 55%, 45% and 22% respectively [21] . Other studies demonstrated that seasonal vaccination had reduced the risk of hospitalization by as much as 79% in diabetics [22] and 54% in persons with COPD [23] . Complications, such as heart attacks in cardiovascular or COPD patients, may be reduced by up to 67% [23] - [26] and the chance of stroke occurrence may be reduced by 24% [27] . Exacerbations of COPD may also be reduced by vaccinations against influenza [28] .

Despite all the known benefits that influenza vaccination brings, the vaccination coverage remains unacceptably low globally. There is some evidence, mostly from Europe and the United States, that seasonal influenza vaccination rates are higher in persons with NCDs than in the general population. However, they still fall short of the WHO [5] and EU [29] recommended target of 75% vaccination coverage. For example, in the United States, only one in three adults with heart disease (34%) received influenza vaccination in 2005 [2] . Similarly, seasonal vaccination coverage in people with chronic NCDs in five European countries in season 2006-2007 ranged from 29.8% in Germany to 59.4% in the UK [30] . Another European study noted that 50% of the unvaccinated eligible people (2 - 64 years old) were those living with NCDs [31] .

Even the fear of an emerging pandemic during the 2009-2010 H1N1 outbreak, did not significantly alter influenza vaccination rates in persons with NCDs. For example, the coverage with an H1N1 vaccine in NCD groups was around 26% in Korea [32] and in the USA [33] and only 21% in France [34] .

Vaccination uptake seems to improve with age. In France, the probability of getting vaccinated was nearly10 times higher in older adults and the chronically ill than in the group of people with chronic NCDs who were less than 65 years old [30] . While vaccine is less effective in preventing influenza in people over 65, it has been shown to significantly reduce complications and deaths in this age group [35] . For example, one Spanish study found that vaccination against influenza in older people with cardiac disease reduced the risk of winter mortality by as much as 37% during four influenza seasons [36] .

Children are a high risk group requiring special attention. Currently, only a few countries recommend the influenza vaccine in healthy children. Children with NCDs, however, should be universally immunized annually. Children’s vulnerability is twofold: first, they are particularly prone to complications from influenza; second, they are also the main vectors of transmitting the infection to family members, who may include older adults and chronically ill persons particularly vulnerable to the disease. However, the data on vaccination coverage of children is limited, and what is available is not particularly encouraging. For example, the Spanish national survey found coverage of 7% and 6.5% among children aged 6 - 23 months and 2 - 15 year olds, respectively [37] . Similarly, pediatric seasonal influenza vaccination has been estimated at between 31% - 66% in Italy [38] and 12% in the United States [39] .

Healthcare workers are another high-risk group in terms of both susceptibility and contagiousness: they are at a risk of exposure to influenza virus since healthcare facilities are a frequent site of nosocomial outbreaks [40] . Healthcare workers can act as vectors as well, unknowingly infecting their patients particularly when the infection is asymptomatic. In addition, they may also have underlying medical conditions, thus increasing the risk of developing a severe disease with complications. Despite the evidence of effectiveness and consistent recommendations by WHO and public health authorities for universal vaccination of healthcare workers, the vaccination coverage in this group remains low, ranging from 48% in France to 13.9% in Poland [18] .

Vaccination rates of US healthcare workers are a bit higher than in Europe, but they are still below the recommended universal coverage. According to the US Centers for Disease Control and Prevention (CDC), the seasonal influenza vaccination rates among healthcare workers in 2013 were 58.9% in the United States [41] . The data from other regions are limited, but what is available is consistent with the findings from Europe and the USA. For example, a study from Israel, which surveyed 27 primary care community health centers in Jerusalem, found that the vaccination rate was at 30.2% among healthcare workers [42] . About a third of the respondents noted that their decision not to get vaccinated had been affected by the negative attitude of the media towards vaccinations.

Another study of the factors influencing healthcare workers’ uptake of influenza vaccination noted that high perceived risk for influenza infection, reduction in the risk to infect patients, knowledge of and agreement with national guidelines, social influence of people close to the respondents, media interest and the ethical perspective that all healthcare workers should get vaccinated, were important considerations whether or not to get a vaccine shot [43] . In general, higher awareness and knowledge of scientific evidence as well as positive attitude towards vaccinations are predictors of a higher vaccination uptake [44] .

Despite national and international recommendations to increase influenza vaccination coverage, particularly in high-risk groups, the coverage is still very low globally. Implementing a targeted-group strategy is complex, and a multipronged approach is needed at both global and country levels to increase vaccination rates.

Influenza vaccination campaigns based on educational interventions influence individuals’ decision to get vaccinated when they understand the risks associated with influenza and are motivated to protect themselves against the infection [30] . Palache, 2011 [62] found that wide-reaching communication campaigns were strongly associated with increased influenza immunization coverage.

In the United States, the US CDC conducts an annual awareness campaign, which includes advertisements, public service announcements and a broad range of free communication materials in a variety of formats (e.g. posters, information sheets, web tools and audio/visual materials) targeted at a range of risk groups [63] . In Brazil, a national campaign provides resources, including radio segments, video adverts, posters and artwork for promotional T-shirts and baseball caps [64] .

Ompad et al., 2006, [65] found that multicomponent programs were the most successful at vaccinating high- risk individuals. Other studies evaluated vaccination interventions using a variety of approaches. Some used mailing to target the high-risk populations in combination with an educational insert or a brochure [66] or a media campaign [67] . Letters alone increased coverage by 5.33% in comparison to the control group [68] . Others reported strategies to increase coverage among high-risk groups ranging from advertising, provider and patient mailings, registry-based telephone calls and patient and healthcare workers education to targeting needle exchange customers and nurses [66] [69] - [72] .

An efficient method to increase coverage among high-risk individuals reported in the literature is to target venues frequented by high-risk groups, such as primary and tertiary care settings situated in hospitals or in clinics and physician practices [65] . This is particularly important for NCD patients, who tend to visit clinics more frequently. However, in today’s disease-centered health systems, opportunities for NCD prevention in clinical settings may be missing. Given the fact that personal recommendation by a healthcare professional, particularly a family doctor, is the single factor most likely to encourage vaccination, using physicians as role models of healthy behaviors can be an effective approach to improve vaccination uptake among NCD patients [73] .

Healthcare workers’ attitudes and beliefs around influenza vaccination are important determinants of vaccine uptake, therefore vaccination awareness campaigns targeting healthcare workers are necessary. This approach is particularly encouraged in North America and Europe.

The US Healthcare Infectious Control Practices Advisory Committee and the Advisory Committee on Immunization Practices recommended that all healthcare organizations use evidence-based approaches in their campaigns and suggested five categories of components for running successful immunization campaigns among healthcare workers (Table 2).

A systematic review of seasonal influenza vaccination campaigns for healthcare workers in the United States found that campaigns with a combination of education or promotion and improved access to the vaccine yielded greater increases in coverage rates [74] .

The World Medical Associations (WMA) Influenza Immunization Campaign, supported by the International Federation of Pharmaceutical Manufacturers and Associations (IFPMA), provides an example of a successful multicomponent global campaign. The campaign uses a combination of education/promotion and role models to

Source: [74] .

raise awareness of the importance of influenza immunizations among physicians and encourages them to set examples by getting vaccinated annually to protect themselves and their patients.

The WMA campaign is in its second phase, which focuses on indirect protection of patients in high-risk groups through physician vaccinations and counseling. The focus of the campaign is aligned with person-cen- tered medicine, which stresses scientific and ethical considerations for promoting clinical care of the person (the total health of the person not just one condition), for the person (to achieve the best possible outcomes), by the person (doctors acting as role models for adopting healthy behaviors) and with the person (physician and patient working together as a team to reach the desired health goals) [75] .

The WMA Campaign is global, and its success in terms of increasing vaccination coverage rates among physicians and, indirectly, of their patients, may not be possible to measure. However, the campaign encourages the work of national member associations to promote influenza vaccinations and provides resources for national immunization programs helping translate the global-level commitment into country-level action, where the impact of increased vaccination coverage will be more visible.

Improving influenza vaccination coverage among people living with NCDs is a complex task, and multiple strategies are needed at global and country levels to achieve the goal of 75% vaccination rate among this group.

At the global level, it is critical to include influenza immunization as part of the monitoring framework for NCDs and underscore the vaccine’s importance in secondary prevention of these diseases.

At the country level, strategies should target not only those at high risk of influenza complications, such as NCD patients, but also those at elevated risk of both contracting and transmitting the virus, such as schoolchildren and healthcare workers.

The general lack of awareness by healthcare workers and particularly clinicians of the explicit relationship between influenza vaccinations and NCD management may be an important impeding factor in vaccination uptake among the vulnerable groups, especially people living with NCDs. Therefore, awareness building campaigns with a strong educational component is key to a better compliance of healthcare workers with current influenza vaccination recommendations.

To design effective vaccination programs globally, more data are needed on the epidemiology of the disease and vaccine uptake rates in vulnerable groups from low- and middle-income countries. What is currently available, comes mainly from countries in Europe and North America and is not sufficient to design strategies that are appropriate for countries’ level of development, resources available for healthcare systems, and historically established immunization practices.

More studies are needed as well to examine socioeconomic status and vaccination uptake. There is a substantial gap in the literature with respect to reporting vaccination coverage by race/ethnicity and socioeconomic status even in North America and Europe. Given the fact that NCDs are on the rise in low- and middle-income countries and tend to concentrate in poorer populations, documenting disparities in vaccination is critical to address NCDs.

Routine surveillance of influenza manifestation and effectiveness of influenza vaccination among people with NCDs and other high risk groups should be performed and shared across all countries to improve vaccine development and coverage rates.

Global and local influenza information and education campaigns targeting the high-risk groups, including healthcare workers, is an effective strategy and should be explored around the world. However, a one-size-fits- all approach would not work. More research is needed to determine the appropriate design and components of influenza vaccination campaigns to make them context-specific and impactful.

We thank Dr Mukesh Haikerwal, Chair of the WMA Council, for reading the manuscript and providing his expert advice. Our thanks are also extended to Shawn Gilchrist, President of Gilchrist Consulting Services, Inc. who contributed to the development of the manuscript and provided background specialist research. We are also grateful to Ms Marie-Cecile Levant for her valuable comments on the early drafts of the manuscript.