Skip Navigation


Journal of Public Health Advance Access originally published online on February 7, 2008
Journal of Public Health 2008 30(1):91-98; doi:10.1093/pubmed/fdn003
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
30/1/91    most recent
fdn003v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Elliot, A. J.
Right arrow Articles by Fleming, D. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Elliot, A. J.
Right arrow Articles by Fleming, D. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2008, Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved

Acute respiratory infections and winter pressures on hospital admissions in England and Wales 1990–2005



Alex J. Elliot, Primary Care Scientist

Kenneth W. Cross, Statistician

Douglas M. Fleming, Director
Birmingham Research Unit of the Royal College of General Practitioners, Lordswood House, 54 Lordswood Road, Harborne, Birmingham B17 9DB, UK

Address correspondence to Alex J. Elliot, E-mail: aelliot{at}rcgpbhamresunit.nhs.uk, alex.elliot{at}hpa.org.uk

Background Hospitals experience winter surges in admissions due to respiratory infections. The roles of acute bronchitis and influenza-like illness (ILI) in the timing and severity of these surges are examined over the years 1990–91 to 2004–05.

Methods Respiratory admissions of persons aged ≥65 years in England and Wales were analysed in relation to patients with ILI or acute bronchitis diagnosed by community-based general practitioners from a sentinel surveillance network.

Results Acute bronchitis and ILI accounted for 46 and 7% of the variation in respiratory admissions, respectively: when admissions were lagged by 1 week, these estimates were 20 and 14%, respectively. Admissions peaked in weeks 52, 01 or 02 (late December to early January) in 14 of the 15 winters. Acute bronchitis peaked during weeks 01 or 02; ILI exhibited greater variability and peaks ranged from weeks 46 (mid-November) to 07 (mid-February). During winters where acute bronchitis and ILI peaked concurrently, surges on hospitals were most severe.

Conclusions During each winter acute bronchitis provides a consistent and major contribution to the winter admissions surge in the elderly. The variable incidence of ILI can increase the surge in admissions, especially when ILI and acute bronchitis peak together.

Keywords: acute bronchitis, general practitioner, hospital admissions, influenza-like illness, sentinel surveillance, respiratory syncytial virus


   Background
TOP
 Background
Methods
 Results
 Conclusions
 Limitations of this study
 Funding
 Acknowledgements
 References
 
Acute respiratory infections caused by viral pathogens are responsible for causing significant levels of morbidity and mortality on a seasonal basis. Of the viruses most commonly associated with seasonal outbreaks of respiratory disease in the community, influenza and respiratory syncytial virus (RSV) have the strongest association,1 although in recent years, other novel pathogens, e.g. human metapneumovirus and human bocavirus have also been implicated.2,3 Influenza and RSV are the main underlying causes of the two most important respiratory syndromes diagnosed in general practice; influenza-like illness (ILI) and acute bronchitis, respectively.1,4–7 More significantly, these two diagnoses represent the most important and commonly diagnosed respiratory diseases in the youngest and oldest groups of the age spectrum. The elderly, in particular, impose considerable burdens on hospital services because of increased admissions to hospital and prolonged bed stay.8

Winter pressures in hospitals are not a new phenomenon but they reached a new level in the UK during the millennium winter (1999–2000), producing major crises in many hospitals which were unable to meet demand.9,10 Traditionally these increases have been associated with acute respiratory infections, in particular those caused by influenza viruses, although it is thought that RSV is an important factor in determining this burden.4,11,12 Winter surges in hospital respiratory admissions in the elderly have been shown to coincide with admissions for cardiovascular disorders.13 These surges have also been shown to peak concurrently with mortality.14–16 It has previously been implied that peaks of influenza-associated hospital admissions and deaths lag 1–2 weeks behind general practitioner (GP) data.17,18

The aim of this study was to investigate the burden of respiratory infections in the elderly on hospital admissions for respiratory disease during the winter. We examined trends of two of the most commonly reported respiratory tract infections in general practice, i.e. ILI and acute bronchitis, and compared the seasonal incidences of these two diagnoses in the elderly with respiratory admissions over a 15 year period (1990–91 to 2004–05).


   Methods
TOP
 Background
 Methods
Results
 Conclusions
 Limitations of this study
 Funding
 Acknowledgements
 References
 
Clinical incidence data
Clinical diagnostic incidence data were recorded by community-based GPs working in the Royal College of General Practitioners Weekly Returns Service (RCGP WRS).19,20 The WRS sentinel surveillance network has consistently reported from a population representative of the national population by age and sex. Bias in the socioeconomic distribution around the year 2001 resulted in a recruitment drive in underrepresented areas.21,22

New episodes of illness diagnosed as ILI (ICD9 487) and acute bronchitis (ICD9 466 & 490) were used to calculate weekly incidence rates per 100 000 population for the age group ≥65 years.

Some minor adjustments were made to these weekly rates. First, where a year contained a week 53 (1992, 1998, 2004), that week was renamed week 1, and all subsequent weeks adjusted up to week 26, which was removed. Second, during the Christmas and New Year, reduced opening hours of GP surgeries and reduced attendances of registered patients during this period cause small fluctuations to the recording of clinical morbidity data. To compensate for this potential bias, the affected week of each year was adjusted by calculating an average of the values for the adjacent weeks.

Hospital admissions data
Hospital admissions data were obtained from Hospital Episode Statistics (HES) for respiratory admissions (ICD9 Chapter VII; ICD10 Chapter J00–J99) in the age group ≥65 years and weekly admission rates per 100 000 calculated using the mid-year populations for England.23 Adjustments were made to the rates for the last 2 or 3 weeks of a financial year, where relatively low values were observed, which we believed to be due to deficiencies in recording. Therefore, data were adjusted for these affected weeks on the basis of averages of adjacent weeks.

Analysis
The associations between the weekly incidence rates of ILI, acute bronchitis and hospital admissions were assessed by performing a multiple time series linear regression after first detrending each series and adjusting for any autocorrelation using the Cochrane-Orcutt transformation (performed using STATA v9.2).24 This regression analysis was repeated using +1, +2 and +3 week lags in the respiratory hospital admissions data.

Attention was then focused upon the three time series over the 21-week period covering the weeks 44–12 (i.e. November to March inclusive) for each winter separately: this period encompasses the seasonal excess of clinical activity recorded during each winter. For each winter, the three series were first plotted and then the associations between weekly variations in admissions with those of ILI and of acute bronchitis examined.


   Results
TOP
 Background
 Methods
 Results
Conclusions
 Limitations of this study
 Funding
 Acknowledgements
 References
 
Secular trends of ILI, acute bronchitis and respiratory admissions
There was an increasing linear trend for respiratory admissions between 1990 and 2005 (Fig. 1). A slight decreasing linear trend was evident for ILI between 1990 and 1999 and for acute bronchitis from the end of 1994 until the end of 1999; and there was no evidence of trend in either series over the last 5 years of the study.


Figure 1
View larger version (32K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Incidence rates of respiratory admissions, ILI and acute bronchitis in the elderly (≥65 years); note different scales on graphs.

 
The relationship between respiratory admissions, ILI and acute bronchitis over 15 years
Multiple time series regression analysis was performed to assess the proportion of respiratory admissions that were likely due to acute bronchitis and ILI. The adjusted R2 indicated that 52% of the variation in respiratory admissions could be attributed to ILI and acute bronchitis. Using this model, we estimated that 7% of respiratory admissions could be attributed to ILI, and 46% to acute bronchitis, over the 15-year period. The relative weekly contributions to admissions varied, up to a maximum of 66% for acute bronchitis (1992–93) and 33% for ILI (1999–2000). We repeated the time series regression with lags of +1, +2 and +3 weeks in the hospital admissions data. Using the +1 week lag model we obtained a similar fit (adjusted R2) to the zero lag model, which accounted for 52% of variation in admissions. However, the +1 week lag model gave new estimates of 14 and 20% of respiratory admissions attributed to ILI and acute bronchitis, respectively (Fig. 2). The regression models using +2 and +3 week lags in respiratory admissions both gave very poor fits to the data.


Figure 2
View larger version (37K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Comparison of the observed rate of respiratory admissions per 100 000 (≥65 years) with the corresponding rates estimated from ILI, acute bronchitis and other (unknown) causes using the regression model for (a) zero lag and (b) respiratory admissions lagged by 1 week.

 
Analysis of weekly data by individual winter periods
For each winter period (weeks 44–12; November to March) the peak week and peak value for respiratory admissions, ILI and acute bronchitis was recorded (Table 1). In 12 out of the 15 winters, respiratory admissions peaked in week 1 or 2, i.e. the beginning of January of each year. Peak values for respiratory admissions ranged from 62 per 100 000 in 1990–91 to 180 per 100 000 in 1999–2000. Acute bronchitis peaked consistently during the same weeks of each winter. The peak values of acute bronchitis ranged from the lowest, recorded during 2000–01 at 432 per 100 000 to the highest, recorded in the winter of 1999–00 at 1061 per 100 000. The weeks during which ILI peaked were far more variable: the earliest influenza epidemic peaked during mid-November (week 46, 1993–94) and the latest activity peaked during late February/early March (week 7, 1997–98). The peak value of ILI recorded was 366 per 100 000 (week 02, 1999–2000) and the lowest 24 per 100 000 (week 03, 2000–01).


View this table:
[in this window]
[in a new window]

  		Table 1 Peak week and peak incidence rate per 100 000 for respiratory admissions, ILI and acute bronchitis in the elderly (≥65 years)

 
The combined affect of ILI and acute bronchitis on respiratory admissions
We aimed to determine whether hospital admissions were accentuated during those years where ILI and acute bronchitis peaked concurrently (within ± 1 week). We categorized winters into those where clinical peaks of acute bronchitis and ILI occurred together, and those where there was separate activity: due to the consistent nature of acute bronchitis, this analysis mainly focused on categorizing ILI activity. Nine winters were selected where acute bronchitis and ILI peaked concurrently (Table 1). The magnitude of the ILI peak was then considered. Those winters in which the incidences of ILI were below the conventionally accepted threshold25 (1990–91, 1992–93, 2002–03, 2004–05) were excluded from the analysis. During the remaining five winters (1991–92, 1995–96, 1996–97, 1998–99, 1999–2000) ILI activity was mainly contained within the surge period, where the maximum proportion of ILI occurred during weeks 52–03, compared to the whole winter (weeks 44–12).

We hypothesized that the greatest winter pressure would be exerted when admissions peaked at a high value with a unimodal distribution (i.e. a leptokurtic admissions curve), rather than in a year when the principle peak of a bimodal distribution of respiratory admissions was of lower magnitude (i.e. a platykurtic curve; Fig. 3). To test this hypothesis we examined the ratio of the peak value of admissions to the mean weekly value of the baseline (non-surge weeks). The resulting peak ratio provided an indication of the acuteness of the curve, and therefore some measure of the surge pressure on the hospital. There was a range of peak rate ratios, ranking from the highest ratio recorded for the winter of 1999–2000 to the lowest in 1997–98.


Figure 3
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Examples of winters where ILI and acute bronchitis co-circulate resulting in a unimodal distribution of respiratory admissions e.g. 1996–97, and where there are two distinct periods of circulation resulting in a bimodal distribution of admissions e.g. 1997–98.

 
The highest peak rate ratios for admissions occurred in those years where we had confirmed concurrent circulation of ILI and acute bronchitis; these were 1.99 for 1991–92, 1.97 for 1995–96, 2.83 for 1996–97, 2.98 for 1998–99 and 3.88 for 1999–2000. Ratios for the remaining years varied between 1.30 and 1.75. We calculated the median value for each categorized group and estimated the difference (1.22) between the two population medians by the median of all the 50 (5 x 10) differences. The confidence interval for the difference between the two medians was also derived from these 50 differences using a method based on the Wilcoxon two sample rank sum statistic.26 The 95% confidence interval of the difference (1.22) between the medians was 0.34–2.20 which indicated that there was a significant difference in the peak value ratio of admissions between those winters where ILI and acute bronchitis circulated concurrently and those where they were separate.


   Conclusions
TOP
 Background
 Methods
 Results
 Conclusions
Limitations of this study
 Funding
 Acknowledgements
 References
 
Main findings of this study
We have examined the relationship between weekly respiratory hospital admissions and the weekly incidence of new cases of ILI and acute bronchitis in the elderly. We found that in 14 out of the 15 years studied, the peak annual surge period for respiratory admissions occurred through weeks 52–02. One exception was the winter of 1993–94; an early and moderately severe epidemic of influenza occurred during November 1993 causing an increase in respiratory admissions that peaked during week 48.27 Otherwise, the peaks were consistently concurrent with peaks in acute bronchitis suggesting a very close association that has previously received little attention. We conclude that acute bronchitis forms a ‘fixed’ burden of admissions in each winter.

In contrast, ILI activity varies from winter to winter; in some winters we detected secondary admission peaks occurring outside the main surge period, which could be attributed to heightened influenza circulation occurring early (e.g. 1993–94) or late in the winter (e.g. 1997–98). In those years where ILI peaked at the same time as acute bronchitis, the admission peaks were higher than in those years when ILI and acute bronchitis peaked at different times. Using a peak rate ratio we were able to assess the distribution of admissions surrounding the peak week and the relative severity of the surge pressure. Ranking these ratios highlighted those winters where there had been well documented pressure on the NHS, e.g. 1998–99 and 1999–2000.9,28,29

When we repeated our regression analysis using the 1 week lag, the relative contributions of ILI and acute bronchitis were different. Although these revised estimates reduce the contribution of acute bronchitis and increase that of ILI, they still confirm and emphasize that acute bronchitis provides the greater burden. The Nguyen-Van-Tam study similarly found a closer relationship between ILI and influenza and pneumonia admissions using a 1 week lag.18 Our results suggest that in patients presenting with ILI and acute bronchitis there are differences in the timing and severity of symptoms at the initial GP consultation.

What is already known on this topic
In this study, our principal analyses have been concerned with weekly data for the months of November to March inclusive which represent the winter period in the UK. This approach seemed more relevant to analysing annual datasets when studying diseases/pathogens that predominantly circulate in winter months. Furthermore, we have analysed weekly new episodes of acute bronchitis in the elderly; little attention has hitherto focused upon this condition and its possible association with respiratory hospital admissions. One limitation to our study arises because, unlike ILI, it is not possible to link the incidence of acute bronchitis in the elderly directly to a virus (e.g. RSV). We have previously shown that acute bronchitis in infants under 5 years of age is very strongly associated with the circulation of RSV, but the findings for patients ≥65 years were ambiguous as to whether it was related to RSV or to the influenza virus.4,30 One study of particular relevance to these findings is that of Mangtani et al.11 They made an analysis of emergency hospitalizations for respiratory disease in relation to surveillance data on RSV and influenza and found that RSV infection was an important determinant of respiratory admissions in infants but appeared less certain for older persons. The reason for these uncertain results in the elderly is the fact that routine RSV counts are predominantly made on infants and not on patients of other age groups.

Despite differences in methodologies, it is important to note a study by Falsey and colleagues, who specifically evaluated RSV and influenza infections in the elderly.31 They noted that hospitalizations for RSV infection were higher in those years where the dominant influenza A subtype was H1N1; in H3N2 dominant years, influenza hospitalizations were two to three times those for RSV infection. Although we have not specifically studied the effect of virus subtype in this study, it is interesting that in the 5 years where surge pressures were highest, influenza H3N2 subtypes were dominant in the community. This warrants further work to determine the effect of influenza (and RSV) subtype on respiratory admissions to determine whether a prediction on the severity of the surge can be assessed from circulating virus strains.

What this study adds
Our findings are unique for two reasons: first, we have studied the surge pressure on hospitals, which for policy makers and health care planners is the critical annual period when resources are stretched; second, we have analysed the clinical diagnoses that are directly responsible for winter surge pressures on hospitals. Other studies have utilized influenza and RSV laboratory reports, however these two pathogens do not cause the entirety of winter respiratory illness and therefore make an underestimation of the burden placed on hospitals. We have used community-based clinical morbidity data which are not based upon laboratory confirmed findings and therefore encompass a wide range of respiratory entities.

It is notable that during 2 years (1998–99 and 1999–2000), winter respiratory admissions peaked 1 week earlier than the GP episodes. These two winters provided the highest peak ratio values, and some of the highest rates of ILI and acute bronchitis recorded over the 15 years. The severity of disease experienced and limited availability of GP services over the holiday period are factors that might have played a role in these findings. However, we must also acknowledge that to compensate for this phenomenon clinical incidence rates were averaged over this period and therefore have to be interpreted with some caution.

Although we have identified the most likely weeks of maximum pressure on hospitals due to respiratory illnesses, we have not attempted to predict the maximum weekly value of admissions. It would be relatively simple to provide an early warning mechanism for hospitals as an alert for the risk of high surge pressure; the RCGP WRS routinely monitors the age-specific incidence of ILI and acute bronchitis and it would be realistic to co-analyse the rates of both conditions in the elderly with a view to raising an alert if both indices appear to be rising concurrently.32 This would provide a window of 2 or 3 weeks prior to the peak of admissions (possibly reduced to 1 week in severe situations such as 1998–99 and 1999–2000), which would provide valuable time for policymakers in secondary healthcare facilities to reallocate resources in preparation for the surge.

Using the data analysed during this study we were able to estimate a crude winter baseline threshold for respiratory admissions; this varied within the range of 40–50 admissions per 100 000 aged ≥65 years. We think it would be advantageous if accurate admission thresholds were formulated to provide hospitals with a basis for measuring the severity of the winter pressure. Clinical thresholds are currently utilized to monitor the activity of ILI circulating the community; incidence rates less than 30 per 100 000 are commonplace in winter; between 30 and 100 are usual when influenza viruses are circulating; rates exceeding 100 represent above average influenza activity and above 200 are exceptional.25 These thresholds are used by the Department of Health as a trigger for widespread use of antiviral drugs by GPs for treating seasonal influenza.33 A similar system for alerting hospitals of imminent surges in admissions may afford more time to prepare for the next serious event.

This study further highlights the need for good quality microbiological testing in the elderly to determine the underlying causes of respiratory disease. The central problem revolves around sampling; we have previously shown that in general practice elderly patients present with respiratory symptoms ~3 days later than children.34 Other factors such as shorter periods of viral shedding, lower viral titres and dry mucosa make viral diagnosis in the elderly difficult compared to diagnosis in children using standard methods.35


   Limitations of this study
TOP
 Background
 Methods
 Results
 Conclusions
 Limitations of this study
Funding
 Acknowledgements
 References
 
A limitation of using data generated from general practice is the coincidence of the peak in respiratory episodes with the disruption in recording due to the closure of surgeries, and reduced attendance over the Christmas/New Year period. Public holidays are always associated with a fall in consulting behaviour and therefore we had to make adjustments to the weeks affected to prevent an underestimation of the burden. Similarly, the apparent fall in admissions recorded at the end of each financial year required minor adjustment.

The GP diagnoses and hospital admissions used in this study are not directly linked and therefore constitute a potential limitation. Although in the WRS data are routinely collected on referrals, these referrals are not linked to episodes diagnosed in individual patients. Additionally, the entry of referral data are often not timely, i.e. they do not appear within the contemporaneous weekly return. We and others have previously utilized GP-based incidence data and hospital admissions data to examine the impact of communicable diseases on primary and secondary health care resources.8,13,36–38 In particular, we have previously studied WRS GP episodes and hospital admissions for asthma; the congruence of temporal variations of these episodes and admissions (particularly in children) was remarkable.39

The incidence of GP-diagnosed ILI in the community has been steadily falling for the last 20 years.19 Since 1999–2000, the incidence of ILI has been relatively low with virtually no secular trend: as a result, the clinical ILI baseline threshold was reduced from 50 per 100 000 to 30 per 100 000.25 This represents a major limitation for studying ILI; low incidence results in indeterminate peaks of ILI making it difficult to determine periods of activity. In this study, we used ILI data collected from other age groups to confirm the results of the ≥65 years age group. We also used laboratory reports for influenza and RSV as a confirmatory process.

This relatively low incidence of ILI and of acute bronchitis since 1999–2000 may be partly due to an intensive media campaign made during that winter which discouraged people with acute respiratory illnesses from consulting their GPs in the expectation of receiving an antibiotic.40 Another contributing factor has been the gradual decline of GP commitment to out-of-hours responsibilities. Though data from deputizing services are incorporated into the practice information system, the delays involved result in these data not being included in the contemporaneous weekly return.


   Funding
TOP
 Background
 Methods
 Results
 Conclusions
 Limitations of this study
 Funding
Acknowledgements
 References
 
The Birmingham Research Unit of the Royal College of General Practitioners is funded by the Department of Health. A.J.E. is jointly funded by the Royal College of General Practitioners and the Health Protection Agency. D.M.F. has received financial support from the pharmaceutical industry in matters relating to influenza and RSV prevention and treatment. A.J.E. and K.W.C. declare no competing interests.


   Acknowledgements
TOP
 Background
 Methods
 Results
 Conclusions
 Limitations of this study
 Funding
 Acknowledgements
References
 
We gratefully acknowledge the contribution of the WRS sentinel practices and their staff to providing the GP episode data; we thank Duncan Cooper (Health Protection Agency West Midlands) for helpful advice on statistical methods.


   References
TOP
 Background
 Methods
 Results
 Conclusions
 Limitations of this study
 Funding
 Acknowledgements
 References
 

  1. Zambon MC, Stockton JD, Clewley JP, et al. Contribution of influenza and respiratory syncytial virus to community cases of influenza-like illness: an observational study. Lancet (2001) 358:1410–6.[CrossRef][Web of Science][Medline]
  2. Manning A, Russell V, Eastick K, et al. Epidemiological profile and clinical associations of human bocavirus and other human parvoviruses. J Infect Dis (2006) 194:1283–90.[CrossRef][Medline]
  3. Stockton J, Stephenson I, Fleming D, et al. Human metapneumovirus as a cause of community-acquired respiratory illness. Emerg Infect Dis (2002) 8:897–901.[Web of Science][Medline]
  4. Fleming DM, Elliot AJ, Cross KW. Morbidity profiles of patients consulting during influenza and respiratory syncytial virus active periods. Epidemiol Infect (2007) 135:1099–108.[Medline]
  5. Monto AS, Gravenstein S, Elliott M, et al. Clinical signs and symptoms predicting influenza infection. Arch Intern Med (2000) 160:3243–7.[Abstract/Free Full Text]
  6. Rezza G, Valdarchi C, Puzelli S, et al. Respiratory viruses and influenza-like illness: a survey in the area of Rome, winter 2004–2005. Euro Surveill (2006) 11:251–3.[Medline]
  7. Wallace LA, Collins TC, Douglas JD, et al. Virological surveillance of influenza-like illness in the community using PCR and serology. J Clin Virol (2004) 31:40–5.[CrossRef][Medline]
  8. Fleming D, Harcourt S, Smith G. Influenza and adult hospital admissions for respiratory conditions in England 1989–2001. Commun Dis Public Health (2003) 6:231–7.[Medline]
  9. Anon. Flu Pushes NHS to Breaking Point. http://news.bbc.co.uk/1/hi/health/595539.stm (20 November 2007, date last accessed).
  10. Goddard NL, Joseph CA, Zambon M, et al. Influenza surveillance in England and Wales: October 1999 to May 2000. Commun Dis Public Health (2000) 3:261–6.[Medline]
  11. Mangtani P, Hajat S, Kovats S, et al. The association of respiratory syncytial virus infection and influenza with emergency admissions for respiratory disease in London: an analysis of routine surveillance data. Clin Infect Dis (2006) 42:640–6.[CrossRef][Medline]
  12. Thompson WW, Shay DK, Weintraub E, et al. Mortality associated with influenza and respiratory syncytial virus in the United States. JAMA (2003) 289:179–86.[Abstract/Free Full Text]
  13. Fleming DM, Cross KW, Pannell RS. Influenza and its relationship to circulatory disorders. Epidemiol Infect (2005) 133:255–62.[CrossRef][Medline]
  14. Ashley J, Smith T, Dunnell K. Deaths in Great Britain associated with the influenza epidemic of 1989–90. Popul Trends (1991) 65:16–20.
  15. Perrotta DM, Decker M, Glezen WP. Acute respiratory disease hospitalizations as a measure of impact of epidemic influenza. Am J Epidemiol (1985) 122:468–76.[Abstract/Free Full Text]
  16. Tillett HE, Smith JW, Clifford RE. Excess morbidity and mortality associated with influenza in England and Wales. Lancet (1980) 1:793–5.[Medline]
  17. Ahmed AE, Nicholson KG, Nguyen-Van-Tam JS. Reduction in mortality associated with influenza vaccine during 1989–90 epidemic. Lancet (1995) 346:591–5.[CrossRef][Web of Science][Medline]
  18. Nguyen-Van-Tam JS, Brockway CR, Pearson JC, et al. Excess hospital admissions for pneumonia and influenza in persons > or = 65 years associated with influenza epidemics in three English health districts: 1987–95. Epidemiol Infect (2001) 126:71–9.[Medline]
  19. Elliot AJ, Fleming DM. Surveillance of influenza-like illness in England and Wales during 1966–2006. Euro Surveill (2006) 11:249–50.[Medline]
  20. Fleming DM. Weekly Returns Service of the Royal College of General Practitioners. Commun Dis Public Health (1999) 2:96–100.[Medline]
  21. Birmingham Research Unit of the Royal College of General Practitioners. Annual Report of the Weekly Returns Service 2006. http://www.rcgp.org.uk/continuing_the_gp_journey/bru/annual_reports.aspx (20 November 2007, date last accessed).
  22. Harcourt SE, Edwards DE, Fleming DM, et al. How representative is the population covered by the RCGP spotter practice scheme? Using Geographical Information Systems to assess. J Public Health (Oxf) (2004) 26:88–94.[CrossRef][Medline]
  23. National Statistics Online. Population Estimates for UK, England and Wales, Scotland and Northern Ireland. (20 November 2007, date last accessed). http://www.statistics.gov.uk/statbase/Product.asp?vlnk=601&More=N.
  24. Intercooled Stata version 9.2 [program]. College Station, TX 77845, USA: Stata Corporation. (2007).
  25. Goddard NL, Kyncl J, Watson JM. Appropriateness of thresholds currently used to describe influenza activity in England. Commun Dis Public Health (2003) 6:238–45.[Medline]
  26. Campbell MJ, Gardner MJ. Calculating confidence intervals for some non-parametric analyses. In: Statistics with Confidence: Confidence Intervals and Statistical Guidelines—Gardner MJ, Altman DG, eds. (1989) London: British Medical Association. 71–9.
  27. Dedman DJ, Joseph CA, Chakraverty P, et al. Influenza surveillance, England and Wales: October 1993 to June 1994. Commun Dis Rep CDR Rev (1994) 4:R164–8.[Medline]
  28. Anon. NHS Struggles to Cope as Flu Crisis Deepens. http://news.bbc.co.uk/1/hi/health/250024.stm (20 November 2007, date last accessed).
  29. Christophersen O. Mortality during the 1996/7 winter. Popul Trends (1997) 90:11–7.[Medline]
  30. Fleming DM, Pannell RS, Elliot AJ, et al. Respiratory illness associated with influenza and respiratory syncytial virus infection. Arch Dis Child (2005) 90:741–6.[Abstract/Free Full Text]
  31. Falsey AR, Hennessey PA, Formica MA, et al. Respiratory syncytial virus infection in elderly and high-risk adults. N Engl J Med (2005) 352:1749–59.[Abstract/Free Full Text]
  32. Birmingham Research Unit of the Royal College of General Practitioners. The RCGP Communicable and Respiratory Disease Report. http://www.rcgp.org.uk/continuing_the_gp_journey/bru/weekly_data.aspx (20 November 2007, date last accessed).
  33. Salisbury DM. Influenza Season 2005/06 – Use of Oseltamivir. http://www.dh.gov.uk/en/Publicationsandstatistics/Lettersandcirculars/Dearcolleagueletters/DH_4128051 (20 November 2007, date last accessed).
  34. Ross AM, Kai J, Salter R, et al. Presentation with influenza-like illness in general practice: implications for use of neuraminidase inhibitors. Commun Dis Public Health (2000) 3:256–60.[Medline]
  35. Mlinaric-Galinovic G, Falsey AR, Walsh EE. Respiratory syncytial virus infection in the elderly. Eur J Clin Microbiol Infect Dis (1996) 15:777–81.[CrossRef][Web of Science][Medline]
  36. Edmunds WJ, Brisson M, Rose JD. The epidemiology of herpes zoster and potential cost-effectiveness of vaccination in England and Wales. Vaccine (2001) 19:3076–90.[CrossRef][Web of Science][Medline]
  37. Harris JP, Jit M, Cooper D, et al. Evaluating rotavirus vaccination in England and Wales. Part I. Estimating the burden of disease. Vaccine (2007) 25:3962–70.[CrossRef][Medline]
  38. Melegaro A, Edmunds WJ, Pebody R, et al. The current burden of pneumococcal disease in England and Wales. J Infect (2006) 52:37–48.[CrossRef][Medline]
  39. Fleming DM, Cross KW, Sunderland R, et al. Comparison of the seasonal patterns of asthma identified in general practitioner episodes, hospital admissions, and deaths. Thorax (2000) 55:662–5.[Abstract/Free Full Text]
  40. Hart M, Livermoore DM, Weinberg JR. Standing Medical Advisory Committee Subcommittee on Antibiotic Resistance. In. The Path of Least Resistance (1998) London: Department of Health. 1–152.

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
30/1/91    most recent
fdn003v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Elliot, A. J.
Right arrow Articles by Fleming, D. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Elliot, A. J.
Right arrow Articles by Fleming, D. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?