Coverages and factors associated with influenza vaccination among subjects with chronic respiratory diseases in Spain

Abstract

Background: The main influenza complications particularly affect patients over the age of 65 years and those with associated chronic diseases, such as respiratory disorders. This study aims to assess vaccination coverage among Spanish children and adults with chronic respiratory diseases (CRDs), and to describe the factors associated with vaccination in 2003. Methods: We analysed 28 113 records of individuals (6869 children and 21 244 adults) included in the Spanish National Health Survey (NHS) for 2003. As a dependent variable, we used the reply (‘yes’ or ‘no’) to the question: ‘Did you have a ‘flu shot in the latest campaign?’ We calculated influenza vaccine coverage as the percentage of individuals with a respiratory disorder (asthma and/or chronic bronchitis and/or emphysema) who reported having been vaccinated against influenza in the most recent campaign. We analysed the influence of sociodemographic, health-status, medical visits and lifestyle variables on vaccination. Results: Vaccination coverage was 19.9% in children and 54.7% in adults suffering CRD. In both age groups, coverages were significantly higher than those observed for non-CRD sufferers. Among adults, older age, not smoking and medical visits to their physician in the preceding 2 weeks were the variables independently and significantly associated with a higher likelihood of receiving the vaccine. Among children, financial factors influence vaccination. Conclusions: Vaccination coverage among children with CRDs is very low. Also among adult CRD sufferers, vaccination remains below the desired level, thereby making it necessary for strategies to be implemented aimed at improving the use of influenza vaccine.

Introduction

Influenza is regarded as a public health problem of the first order.¹ The main complications particularly affect patients over the age of 65 years and those with associated chronic diseases, such as respiratory disorders.² During influenza epidemics, excess morbidity and mortality have been observed across all age groups in individuals with respiratory complications.^3,,4 A study undertaken in 500 US hospitals from 1979 to 2001 reported that a mean of 200 000 individuals were hospitalized each year due to chronic respiratory disorders associated with influenza infection.⁵ Among chronic respiratory diseases (CRDs), asthma and bronchitis have considerable importance owing to their high prevalence.^6,,7

Generally, influenza vaccination programmes have been targeted at those groups of individuals that register the highest risk of complications.⁸ Current observational studies suggest that vaccination substantially reduces hospital stay and mortality in patients––particularly the elderly––with respiratory problems.^9,,10 Furthermore, the vaccine is effective (76%) in preventing this infection in patients with chronic obstructive pulmonary disease (COPD) regardless of its severity.¹¹ Indeed, a study undertaken in Spain reported that influenza vaccination was responsible for bringing about a significant reduction in pneumonia-related hospital admissions in non-institutionalized subjects over the age of 64 years.¹²

As occurs in other European countries and America, in Spain the vaccine is universally indicated in adults over the age of 64 years and in individuals of any age with CRDs.^13–15

According to the international literature, vaccination coverage in asthmatic children is low, ranging from 9 to 25%.^16–18 In Spain, data on hospitalized asthmatic children show that this coverage stands at 28.6%.¹⁹

Despite the fact that the benefit of annual influenza vaccination in subjects with concomitant––including respiratory––diseases is indicated in the literature, in these risk groups, coverages of more than 60% are not achieved among the over-64 age segment.^20,,21 In Spain, the coverages recorded for the adult Spanish population (over 15 years of age) with CRDs are: 44.7% for 1993; 45.6% for 1995–1997 and 48.8% for 2001.²²

The use of health surveys has been recommended as a tool for the purpose of studying vaccination coverages and associated factors in individuals with chronic health problems.^14,,23,24 Accordingly, this study sought to assess vaccination coverage among Spanish children and adults with CRDs, and to describe the factors associated with vaccination in 2003.

Methods

A cross-sectional study was conducted on influenza vaccination coverage in Spain, using the 2003 National Health Survey (NHS). This survey targeted a representative sample of the non-institutionalized Spanish population, using multistage sampling stratified by clusters, with proportional random selection of primary sampling units (towns) and secondary units (sections), and selection of end units (individuals) by random routes and sex and age quotas. Children's questionnaires were completed by parents or tutors. Details of NHS methodology are described elsewhere.²⁵

For study purposes, individuals with associated respiratory disorders (asthma and/or chronic bronchitis and/or emphysema) were selected from the 2003 survey. The answer (‘yes’ or ‘no’) to the question ‘Did you have a ‘flu shot in the latest campaign?’ was used as the dependent variable, and the following were analysed as independent variables: sociodemographic variables (age, gender, income, size of dwelling and educational level); variables linked to the use of healthcare services (medical visit to a physician in the preceding 2 weeks, use of emergency medical services in the preceding year and admission to hospital in the preceding year) and lifestyle––and health-related variables (perception of health status, smoking habit, physical activity and obesity) and presence of other associated chronic diseases (diabetes and/or heart disease). In addition, vaccinated subjects were asked: whether said vaccination had been administered on medical indication or at their own request; and, whether they had been vaccinated at their place of work or study.

Study subjects were classified into two age groups, with ‘child’ being defined as any individual under 18 years of age, and ‘adult’ as any individual aged 18 years or over.

Influenza vaccine coverage was calculated by defining it as the percentage of individuals with a respiratory disorder (asthma and/or chronic bronchitis and/or emphysema) who reported having been vaccinated against influenza in the most recent campaign.

A bivariate analysis was performed for 2003, cross-tabulating the independent variables with the variable of influenza vaccination among subjects with CRD. The statistical Chi-squared method was used for bivariate comparison of proportions, with statistical significance being set at P < 0.05 (P-values are two-tailed).

To estimate the independent effect on receipt of influenza vaccine for each of the age groups, of variables that had shown a significant association in the bivariate analysis, adjusted odds ratios were obtained by means of multivariate logistic regression models.

All estimates were made by incorporating the complex sampling design and weighting factors, using the ‘svy’ (survey commands) functions of the STATA 9.1 programme, which enabled us to incorporate the sampling design into our statistical calculations (proportions, confidence intervals (CIs), Chi-squared, logistic regression).

Results

A total of 28 113 records of individuals (6869 children and 21 244 adults) who answered the NHS questionnaire were analysed. Of these, 1482 individuals (5.2%) reported presenting with asthma and/or bronchitis and/or emphysema, with 49.3% stating that they had been vaccinated.

Shown in table 1 is influenza coverage of children and adults, with a breakdown indicating the presence or absence of CRDs by sociodemographic variables.

Among CRD sufferers, vaccination coverage was 19.9% in children (n = 313) versus 54.7% in adults (n = 1169), but in both age groups coverages were significantly higher than those observed for non-CRD sufferers.

For each age group, coverage was seen to rise with age, with these differences proving statistically significant for subjects over the age of 17 years. Elderly subjects (aged 65 years and over) accounted for the—significantly––highest proportion of influenza vaccination in 2003 (77.8%).

Vaccination coverage was observed to be significantly higher among children whose parents/tutors reported a monthly income of over €600.

Among adults, a significantly higher percentage of men (57.6%) were vaccinated than women. Coverage was higher among subjects without a university education and those reporting a lower monthly income level.

Results for variables linked to health status and use of healthcare services are shown in table 2. Among adults with CRDs, those who perceived their health more negatively were vaccinated to a greater extent than those who considered their health to be good. Likewise, the presence of comorbidity (heart disease and/or diabetes) was associated with significantly higher coverages (72.4% versus 48.1%), and individuals who paid more medical visits to their physician or had been admitted to hospital in the preceding 12 months also registered significantly higher coverages.

Influenza vaccinations according to lifestyle, together with the reason for vaccination are shown in table 3. There was a significantly higher percentage of vaccination among adults with CRDs who were non-smokers than among those who were smokers at the date of study. Among CRD sufferers in both age groups, the reason most cited for having influenza vaccination was medical prescription, with this accounting for over 90%. These percentages were significantly lower among healthy subjects.

Table 4 shows the odds ratios, adjusted by logistic regression, along with their 95% CIs for having received influenza vaccination in the most recent campaign. This table only displays variables in which the statistically significant association was maintained after multivariate adjustment. Following multivariate adjustment, the likelihood of vaccination in children with CRDs continued to be associated with a higher income level among parents.

The variables that were independently and significantly associated with a higher likelihood of receiving the vaccine among adults with CRDs were: older age; not smoking and medical visits to their physician in the preceding 2 weeks.

Discussion

The results of our study show that influenza vaccination coverage among Spanish children with CRDs is very low (19.9%), while in neighbouring countries, such as Italy, this figure is 26.3% in children aged under 14 years.¹⁶ However, in a published study undertaken in a number of US Health Maintenance Organizations, coverage among asthmatic children ranged from 9 to 10% in 2000.¹⁷ In subsequent studies conducted in the USA, 16.7% of children with asthma received an influenza vaccination during the 2001–02 season, and 21.8% received the vaccine during the 2002–03 season.²⁶

Our study shows that after multivariate adjustment, the only factor significantly associated with vaccination among child CRD sufferers was monthly household income above €600, i.e. the children of high-income parents were more likely to be vaccinated. Similar results have been reported in Spain and the United States.^27–29 Some authors have suggested that the explanation for these results may lie in the fact that, having greater economic resources could mean better healthcare, and so more contact with healthcare professionals and a greater probability of vaccination.^30,,31

In our study, influenza vaccination coverage among adults was slightly over 50%. In the USA, coverages have been calculated separately for patients with asthma or chronic bronchitis. Calculations based on the National Health Interview Survey (NHIS) show that 35.1% of asthmatic adults had influenza vaccination in 1999, 36.7% in 2000 and 33.3% in 2001.²⁰ In 2003, median influenza coverage among asthma sufferers aged 18–64 years was 34.0% (range: 22.5–46.6%).³² Insofar as patients with bronchitis are concerned, the 1999 NHIS calculated coverage as being 43%.³³ In Europe, data obtained via telephone surveys in 2004 reflect lower coverage figures in individuals with ‘lung disease’, namely: 8% (95% CI: 3–18%) in Poland; 11% (95% CI: 7–18%) in Sweden and 20% (95% CI: 15–27%) and 30% (95% CI: 22–41%) in Spain.² A study conducted at Spanish health centres observed a coverage of 87% among COPD sufferers over the age of 40 years.³⁴

The progressive increase in influenza coverage with age reported in other studies is in line with our results.^22,,32

The finding that smoking is a negative predictor of influenza vaccination is in line with the results of different studies conducted on high-risk adult and elderly individuals.^34,,35

In our study, adult CRD sufferers reporting at least one medical visit to their physician in the preceding 2 weeks were more likely to be vaccinated. Similar results have been previously described.³⁶ It is only logical to surmise that the higher vaccine coverages found among the most frequent users of healthcare services are attributable to the fact that such subjects have a greater likelihood of receiving recommendations to be vaccinated by their medical practitioners. Indeed, the great importance of medical advice in ensuring influenza vaccination has been highlighted in other studies.³⁷ This is coherent with our results, as the reason most cited by patients for being vaccinated was medical indication. Nevertheless, different authors have reported that health professionals may possibly often waste the opportunity to recommend influenza vaccination to patients deemed to be at risk.^38,,39

Different strategies are mentioned in the literature for improving coverage in high-risk patients, including those with CRDs.^40–43 Physicians and paediatricians play an essential role, which is why it is essential for stress to be continued to be laid on information about the benefits of the vaccine, its effectiveness and safety as a fundamental strategy. Moreover, the literature indicates that other strategies are also effective, and among these, special mention should be made of: lowering the age at which the influenza vaccine recommendation becomes universal; telephoning or mailing personal reminders; compliance monitoring; using computerized systems to identify high-risk patients; improving medical records; empowering nurses to vaccinate patients directly and drawing up purpose-made influenza vaccination timetables.^40–43

Our study has a number of limitations. First, it has to be said that the NHS does not enable any distinctions to be drawn in terms of whether individuals are asthmatic or have chronic bronchitis, the date on which they presented with this diagnosis or the severity of such. Other authors agree with us in not separating by disease but rather analyzing subjects ‘with associated CRDs’.^1,,44

Second, another possible limitation lies in the use of data obtained through self-reports of influenza vaccination. However, in other studies, this type of direct testimony has been compared against the vaccine records kept by physicians, and has shown itself to be highly sensitive.^45–47 Moreover, a good degree of concordance has been observed between the two sources.^45–47 As with influenza vaccination, the validity of the question addressing the presence of CRDs was likewise not assessed, though prevalence, sociodemographic characteristics and health status are similar to those of other Spanish studies, something that speaks in favor of its validity.^6,,7,34

Third, the results obtained are only applicable to the non-institutionalized population, which means that the results cannot be generalized to the total population.

Lastly, the initial response rate to the 2003 NHS was 67%, so that the existence of possible non-response bias must therefore be considered.⁴⁸ Nevertheless, we regard this as being unlikely because the coverages observed are coherent with vaccination estimates based on other sources, such as number of doses distributed (212/1000 population in 2003), according to figures furnished by the Spanish Ministry of Health & Consumer Affairs.⁴⁹ The NHIS carried out periodically in the United States, with a methodology and content similar to the Spanish National Health Survey, is used systematically for the study of vaccination coverage, its trend and associated factors.^{23,,33,50–53}

Conclusions

Pre-eminent among the relevant conclusions to be drawn from our study is the fact that vaccination coverage among children with CRDs is very low, with only 19.9% receiving the vaccine. It is noteworthy that financial factors influence vaccination of children with CRDs in Spain. Among adult CRD sufferers, vaccination remains below the desired level, thereby making it necessary for strategies to be implemented aimed at improving the use of influenza vaccine among CRD patients, especially those aged under 64 years without other chronic conditions.

Acknowledgements

This study forms part of research funded by the FIS (Fondo de Investigaciones Sanitarias-Health Research Fund) grant no. PI060452 from the Instituto de Salud Carlos III.

Conflicts of interest: None declared.

References