Seasonality of Pneumonia Admissions and Its Association with Climate; A Eight-year Nationwide Population-based Study

(1)

SEASONALITY OF PNEUMONIA ADMISSIONS AND ITS

ASSOCIATION WITH CLIMATE: AN EIGHT-YEAR NATIONWIDE POPULATION-BASED STUDY

Hsiu-Chen Lin,1Ching-Chun Lin,2Chin-Shyan Chen,3and Herng-Ching Lin2 1_{Department of Pediatric Infection, Taipei Medical University and Hospital, Taipei,} Taiwan; School of Medicine, Taipei Medical University, Taipei, Taiwan

2_{School of Health Care Administration, Taipei Medical University, Taipei, Taiwan} 3_{Department of Economics, National Taipei University, Taipei, Taiwan}

The aim of the study was to examine seasonal variability in monthly admissions for community-acquired pneumonia (CAP) in Taiwan. Our study sample comprised 477,541 pneumonia patients in Taiwan between 1998 and 2005, inclusive. Results showed a fairly consistent seasonal pattern of pneumonia admissions, regardless of sex and age, and for the groups combined. Seasonal trends showed a peak in hospitaliz-ations from January through April, followed by a sharp decrease in May and a trough from August through October. The auto-regressive integrated moving average (ARIMA) test found significant seasonality for all age and sex groups and for the whole sample (all p< 0.001). After adjusting for seasonality, month, and trends, the ARIMA regression models revealed that the monthly pneumonia admissions rate was signifi-cantly associated with ambient temperature, for the total sample, for female groups, and for the 65–74 and ≥75 age groups (all p < 0.01). A 1°C decrease in ambient temp-erature was associated with roughly a 0.03 increase in monthly pneumonia admissions rate (per 10,000 people) for the entire sample. We conclude the monthly pneumonia admissions rate was significantly associated with seasonality, and was higher in periods with low ambient temperatures. (Author correspondence: [email protected]) Keywords Pneumonia, Seasonality, Ambient temperature, Climate

INTRODUCTION

Community-acquired pneumonia (CAP) is a common infectious illness that causes significant morbidity and mortality worldwide, despite improvements in antibiotic and supportive treatment (Bartlett & Mundy, 1995; Bartlett et al., 2000; Lieberman et al., 1996b; Mandell, 1995;

1647

Address correspondence to Herng-Ching Lin, School of Health Care Administration, Taipei Medical University, 250 Wu-Hsing St., Taipei 110, Taiwan; Tel.: 886-2-2776-1661 ext. 3613; Fax: 886-2-2378-9788; E-mail: [email protected]

Submitted May 4, 2009, Returned for revision June 22, 2009, Accepted July 15, 2009 ISSN 0742-0528 print/1525-6073 online

DOI: 10.3109/07420520903520673

Chronobiol Int Downloaded from informahealthcare.com by Taipei Medical University on 05/30/11

(2)

Mandell et al., 2003). The reported incidence of radiographically con-firmed CAP per year in different populations has varied between 1.3 and 11.6 cases/1,000 inhabitants (Almirall et al., 2000; Gutierrez et al., 2006; Jokinen et al., 1993; Macfarlane et al., 1993; Marrie et al., 1989; Marston et al., 1997; Ortqvist et al., 1990; Santos de Unamuno et al., 1998; Viegi et al., 2006; Woodhead et al., 1987). Approximately 20% of CAP patients require hospitalization (Oosterheert et al., 2003), and 10–35% of admitted patients require treatment in an intensive care unit (ICU) (Ewig et al., 1998; Moine et al., 1994). Reported mortality in such patients is between 20 and 50%. Pneumonia is the sixth most common cause of death in the United States (Bartlett et al., 2000). Similarly, it is the fourth leading cause of death in Taiwan (Taiwan Department of Health, 2009), placing a significant burden on the healthcare system.

Like other pulmonary diseases or cerebrovascular accidents (Lee et al., 2008; Lin et al., 2008; Tsai et al., 2007), seasonal variation in hospitaliz-ation rates for CAP (CAP-H) has long been observed by researchers, with the focus of much of the research being on mortality (Donaldson & Keatinge, 2002), specific sub-populations such as the very young (Muller-Pebody et al., 2002) and elderly (Nguyen-Van-Tam et al., 2001), or particular etiological agents (Dowell et al., 2003; Kim et al., 1996; Lieberman et al., 1996a). Studies consistently report distinct seasonal patterns, with peak mortality and morbidity occurring in the winter and troughs in the summer (Carrat & Valleron, 1995; Dowell et al., 2003; Saynajakangas et al., 2001). Variations in the seasonal timing of peaks and troughs by age and sex have also been reported (Crighton et al., 2004; Saynajakangas et al., 2001).

However, previous studies have frequently been characterized by a number of methodological limitations, including short time periods (Dowell et al., 2003; Lieberman et al., 1996a) and small clinical samples (Lieberman et al., 1996a, 1999). We believe that short study periods and small samples do not enable one to determine whether the seasonal varia-bility found for an individual year is incidental or a phenomenon that occurs in other years as well. Moreover, many prior studies have tended to use CAP-H patient samples from selected hospitals, an approach that could clearly leads to selection bias due to variations in practice between hospitals, such as criteria for admissions and number of available beds (Lieberman et al., 1996a, 1999). Furthermore, the majority of popu-lation-based studies have also made no attempt to explore the relation-ship between climatic parameters and CAP-H (Crighton et al., 2004; Saynajakangas et al., 2001). Because previous studies have been mainly conducted in frigid and temperate regions of the world, their results are difficult to generalize to other regions. For example, the climate in tropi-cal and subtropitropi-cal regions has less extreme variation compared to the summer and winter swings in frigid or temperate zones.

(3)

In light of the above, the aim of the study was to examine the seasonal variability of CAP-H in Taiwan, which has a largely subtropical climate, using a time series analysis approach to assess an eight-year nationwide population database. This study set out to investigate the meteorological factors (i.e., ambient temperature, relative humidity, atmospheric pressure, rainfall, and total hours of sunshine) associated with CAP-H, according to age and sex groupings. We used nationwide hospital admissions records from 1998 to 2005, inclusive. To our knowledge, this is the largest and most complete nationwide population-based study to investigate the dependence of CAP-H admissions rates on meteorological conditions in tropical and subtropical regions.

METHODS

Hospitalization Data

This study used hospitalization data from 1998 to 2005 from the Taiwan National Health Insurance Research Database (NHIRD), pub-lished by the Taiwan National Health Research Institute. The NHIRD covers all hospitalization and medical benefit claims for about 98% of Taiwan's population of over 23 million, and it is plausibly the largest and most comprehensive population-based health database in the world. The NHIRD provides one primary diagnosis from the International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) and up to four secondary diagnoses for every hospitalization claim. As these were de-identified secondary data, released to the public for research purposes, after consulting with the director of the IRB of our university, the study was exempt from full review; nonetheless, it conforms to the international ethical standards of the journal (Portaluppi et al., 2008).

Study Sample

We selected all adult (≥18 yrs) inpatient claims between January 1998 and December 2005 with a principal admission diagnosis of pneu-monia (ICD-9-CM codes 480–483.8, 485–486, and 487.0). We excluded readmissions within 30 days, as these were regarded as the same episode. Ultimately, our study sample comprised 477,541 cases of pneumonia patients for 1998–2005, inclusively.

Population Data

This study used population registry data in Taiwan to compute the monthly incidence of pneumonia per 10,000 people for the period from 1998–2005. For this study, we defined the monthly pneumonia incidence as

(4)

the proportion of the total monthly admissions for pneumonia as a fraction of the entire island's population. Population data were obtained from the Population Affairs Administration of the Ministry of the Interior, Taiwan.

Meteorological Data

We used meteorological data, including average monthly ambient temperature, relative humidity, atmospheric pressure, rainfall, hours of sunshine, and maximum and minimum temperatures, from 19 Taiwan Central Weather Bureau (CWB) observation stations. Although the CWB has 26 observation stations across the island, we discarded meteorological data from seven stations located in mountainous regions with very sparse population in order to more closely represent the conditions experienced by the majority of the population. The monthly mean values were then derived by averaging the monthly data from the remaining 19 stations. Because Taiwan is a relatively small island, with a total land area slightly less than 36,200 km2, we could use monthly mean values for climatic data to explore its association with pneumonia admission rates.

Statistical Analysis

We used the Statistical Package for the Social Sciences (SPSS Statistics 17.0 for Windows, 2007, SPSS, Chicago, Illinois, USA) to perform the analysis. Monthly pneumonia admission rates/10,000 people were calcu-lated over the 96-month span and categorized by sex and age groupings (of 18–64, 65–74, and ≥75 yrs). Seasonality is a general feature of time series patterns, so the seasonality of the pneumonia admission rates was evaluated using the ARIMA method (Auto-Regressive Integrated Moving Average), which describes a univariate time series as a function of its past values. In addition, a series of cross-correlations was used to reveal the association between climatic factors and monthly pneumonia admission rates. We also adopted the ARIMA regression method as a means of eval-uating the associations between climatic parameters and monthly pneu-monia admission rates, after adjusting for time-trend effect. The selection of the final parameters was based upon the lowest mean absolute percen-tage error or mean absolute error, allowing us to choose the best model from the family of ARIMA regression models. All p values of< 0.05 were considered statistically significant in this study.

RESULTS

Admission Rates

The number of admissions for pneumonia by year throughout the study period was 47,752 (1998), 55,554 (1999), 56,137 (2000), 60,063

(5)

(2001), 62,620 (2002), 56,795 (2003), 61,046 (2004), and 77,574 (2005). These figures correspond to annual admission rates of 31.58, 36.03, 35.76, 37.60, 38.56, 34.47, 36.53, and 45.84 per 10,000 people, respectively. Throughout the study period, the monthly pneumonia admission rates ranged from a low of 1.91/10,000 in November 1998 to a high of 5.64/ 10,000 in March 2005, with a mean of 3.27/10,000 and a standard devi-ation of 0.85/10,000. The mean monthly pneumonia admission rate for males was 3.95/10,000 and 2.57/10,000 for females. The mean monthly levels for the five climatic variables across the entire 8-yr study period were 23.12°C temperature, 78.43% relative humidity, 998.88 hPa atmospheric pressure, 183.73 mm rainfall, and 158.27 monthly hours of sunshine.

Table 1 summarizes the profile of the sampled cases. The mean length of stay for sampled admissions was 11.41 days (standard deviation ±13.02 days) and mean hospitalization costs were NT$65,548 (standard deviation±NT$113,640). The average exchange rate in 2003 was US$1.00=NT$ 33.50. In total, 60.91% of the admissions were male and 37.86% were≥75 yrs of age. As for co-morbidities or complications, 19.51% had an additional diagnosis of essential hypertension, 15.32% diabetes, and 12.38% acute respiratory failure. In addition, 10.75% of patients admitted for pneumonia were diagnosed with stroke, 8.10%

TABLE 1 Pneumonia inpatients in Taiwan, 1998–2005 (n = 477,541)

Variable n (%)

Sex

Male 290,847 (60.91)

Female 186,694 (39.09)

Age group (yrs)

18–64 193,924 (40.61) 65–74 102,805 (21.53) ≥75 180,812 (37.86) Complications or co-morbidities Essential hypertension 93,168 (19.51) Diabetes 73,159 (15.32)

Acute respiratory failure 59,120 (12.38)

Stroke 51,336 (10.75)

Septicemia 38,681 (8.10)

Chronic airways obstruction, not elsewhere classified 36,627 (7.67)

Renal disease 32,855 (6.88)

Asthma 31,279 (6.55)

Pleurisy 29,130 (6.10)

Chronic bronchitis 25,501 (5.34)

Congestive heart failure 23,208 (4.86)

Anemia 20,916 (4.38)

Pulmonary tuberculosis 18,576 (3.89)

Malignant neoplasm of trachea, bronchus or lung 14,804 (3.10)

Gout 13,276 (2.78)

Fever 7,354 (1.54)

_{Only present complications or co-morbidities occurring in over 1.5% of all cases are presented.}

(6)

septicemia, 7.67% chronic airway obstruction, and 6.88% renal disease. Also, asthma was identified in 6.55% of the patients; 6.10% had pleurisy, 5.34% chronic bronchitis, and 4.86% congestive heart failure. In total, 17.5% of the sampled patients had more than one co-morbidity.

Seasonality

According to the Taiwan CWB, spring occurs in Taiwan from March to May, summer from June to August, autumn from September to November, and winter from December to February. Monthly variations in pneumonia admission rates for each sex group, each age group, and both sex groups combined are presented in Figures 1 and 2. A similar seasonal pattern of pneumonia admissions is apparent for both men and women, each age group, and all groups combined. Hospitalizations were most numerous between January through April, sharply decreased in May, and were least numerous between August through October. Then, an upward trend started in November and peaked again in January. The ARIMA test for seasonality found significance for all age and sex groups and for the whole sample (all p< 0.001).

Climatic Influences

Figure 3 depicts the variations of monthly mean ambient tempera-ture, relative humidity, atmospheric pressure, rainfall, and sunshine hours, together with the corresponding monthly pneumonia admission rates. After adjusting for seasonality, month, and trends, the ARIMA regression models (results in Tables 2 and 3) revealed that the monthly pneumonia admissions rate was significantly associated with ambient

FIGURE 1 Monthly hospitalization rates (per 10,000) for pneumonia in Taiwan, by sex from 1998 to 2005.

(7)

temperature for the entire sample, the female group, and for the 65–74 and≥75 age groups (all p < 0.01). Based on the model parameters in Table 2, we can roughly conclude that a 1°C decrease in ambient temp-erature is associated with a ∼0.03 increase in the monthly pneumonia admissions rate (per 10,000 people) for the whole population.

DISCUSSION

This study used an 8-yr population-based dataset to examine seasonal variations in pneumonia admission rates and explore their association with weather conditions in Taiwan. Our results indicate that the monthly pneumonia admission rate was significantly associated with seasonality for all age and sex groups, and it was also significantly associated with ambient temperature for the total sample, females, and the 65–74 and≥75 age groups.

In the present study, we found that the annual pneumonia admission rate for the year of 2005 was much higher than it was between 1998 and 2004. This could be due to the high rate of influenza occurring in 2005 in Taiwan. According to data released by the Taiwan Centers for Disease Control, the annual consultation rate for influenza-like illnesses reported by general practitioners was noticeably higher in year 2005 compared to previous years (1998 to 2004). We also found that hospitalizations peaked from January to April (late winter and early spring), and a trough appeared from August to October (late summer and early autumn) for all age and sex groups and for the whole sample. Findings from prior studies parallel our own, showing the peak month for hospitalizations for pneu-monia to be January and the fewest admissions in August in southern Israel and England (Lieberman et al., 1996a; Nguyen-Van-Tam et al., 2001). Furthermore, the number of pneumonia admissions in winter and FIGURE 2 Monthly hospitalization rates (per 10,000) for pneumonia in Taiwan, by age, from 1998 to 2005.

(8)

spring was significantly higher than in the summer and fall for all ages and each age group separately in the United States, Finland, Sweden, and France (Carrat & Valleron, 1995; Dowell et al., 2003; Ortqvist et al., 1990; Saynajakangas et al., 2001). The increased presence of circulating patho-gens in the winter months may explain this pattern. Most studies show that the most frequent bacteria identified in pneumonia patients was FIGURE 3 Mean monthly trends of climatic factors.

(9)

Streptococcus pneumoniae (Diaz et al., 2007; Lauderdale et al., 2005). The occurrence of Streptococcus pneumoniae infection is related to seasons. In the Northern Hemisphere, pneumonia usually peaks between December through February. People with weaker immune systems are at higher risk than others for Streptococcus pneumoniae infection; therefore, the monthly pneumonia admission rate is higher in winter.

We also found the monthly pneumonia admissions rate was signifi-cantly higher at times of low ambient temperatures. In recent years, influ-enza viruses have been recognized as a potential common cause of pneumonia in adults (Jennings et al., 2008; Johnstone et al., 2008). One study reported that colder temperatures are associated with increased viral activity in subtropical regions of Brazil (Lowen et al., 2008). Shahid TABLE 2 ARIMA regression analysis by sex group

Sex group

Independent variable

Total Male Female

β SE t-value Β SE t-value β SE t-value

Intercept 20.52 22.58 0.91 23.82 29.97 0.79 21.41 17.15 1.25 Atmospheric pressure −0.02 0.02 −0.73 −0.02 0.03 −0.63 −0.02 0.02 −1.07 Ambient temperature −0.03 0.01 −2.73† −0.16 0.09 −1.78 −0.03 0.01 −3.37† Relative humidity 0.01 0.05 0.28 −0.01 0.09 −0.17 0.03 0.04 0.80 Rainfall 0.01 0.01 1.26 0.01 0.00 1.10 0.01 0.01 1.32 Hours of sunshine 0.01 0.01 0.56 0.01 0.00 0.41 0.01 0.01 0.52 Trend 0.01 0.01 3.71‡ 0.02 0.00 4.65‡ 0.01 0.01 2.48 January 1.18 0.18 6.61‡ 1.20 0.21 5.71‡ 1.14 0.16 7.31‡ February 1.37 0.23 6.05‡ 1.35 0.25 5.34‡ 1.37 0.21 6.65‡ March 1.30 0.25 5.16‡ 1.35 0.28 4.78‡ 1.24 0.23 5.39‡ April 0.78 0.33 2.35 0.89 0.39 2.29 0.67 0.29 2.29 May 0.17 0.43 0.40 0.34 0.51 0.65 0.01 0.36 0.02 June −0.17 0.50 −0.34 −0.04 0.61 −0.06 −0.30 0.42 −0.72 July −0.06 0.56 −0.11 0.16 0.68 0.23 −0.27 0.47 −0.58 August −0.22 0.55 −0.41 −0.03 0.67 −0.05 −0.40 0.46 −0.88 September −0.56 0.48 −1.18 −0.42 0.58 −0.72 −0.70 0.40 −1.75 October −0.56 0.36 −1.54 −0.50 0.44 −1.15 −0.60 0.31 −1.94 November −0.53 0.21 −2.47 −0.53 0.26 −2.06 −0.51 0.18 −2.83† AR1 0.47 0.13 3.63‡ 0.49 0.14 3.43† 0.48 0.12 4.06‡ MA1 0.47 0.13 3.75‡ 0.34 0.15 2.28 0.57 0.11 5.36‡ Akaike info criterion (AIC) 0.94 1.25 0.67 Schwarz criterion (SC) 1.48 1.79 1.20 R2 _0.86 _0.85 _0.87 _p_{< 0.05,}†_p_{< 0.01,}‡_p_{< 0.001}

Selection of the final parameters was based upon the lowest AIC and SC. Reference group= December.

Abbreviations: AR1= autoregressive, lag 1; MA1 = moving average, lag 1.

(10)

et al. (2009) demonstrated that influenza virus lost infectivity after 30 min at 56°C, after 1 day at 28°C, but remained viable for more than 100 days at 4°C. This study shows that influenza viruses are more active in low ambient temperatures, which explains, at least in part, why the monthly pneumonia admission rate is higher in cold weather. In addition, seaso-nal difference in immunity function or crowding may also play a role in seasonality of pneumonia.

Prior studies documented that aging is associated with a higher risk of acquiring pneumonia caused by Streptococcus pneumoniae and influenza virus (Gutierrez et al., 2006; Muller-Pebody et al., 2002; Nguyen-Van-Tam et al., 2001). We found that the pneumonia admission rate of the ≥75 yrs age group was higher than other age groups. Weaker immune systems and the higher incidence of infection can explain the higher admissions rate for the ≥75 yrs age group. The World Health Organization states that vaccines are the only available tool to prevent pneumococcal infections. An increase in vaccine coverage from 50–60% TABLE 3 ARIMA regression analysis by age group

Age group

Independent variable

18–64 yrs 65–74 yrs ≥ 75 yrs

β SE t-value Β SE t-value β SE t-value Intercept 13.22 13.32 0.99 119.91 80.22 1.49 77.20 186.83 0.41 Atmospheric pressure −0.01 0.01 −0.85 −0.11 0.08 −1.34 -0.04 0.18 -0.23 Ambient temperature −0.05 0.04 −1.24 −0.02 0.01 −3.22† -0.04 0.01 -3.06† Relative humidity 0.03 0.03 1.00 −0.14 0.15 −0.91 -0.07 0.38 -0.19 Rainfall 0.01 0.01 1.40 0.01 0.01 0.57 0.01 0.01 0.93 Hours of sunshine 0.01 0.01 0.15 0.01 0.01 0.90 0.01 0.01 0.32 Trend 0.01 0.01 2.31 0.01 0.01 0.69 0.08 0.28 2.99† January 0.67 0.10 6.50‡ 1.96 0.54 3.64‡ 7.86 1.41 5.59‡ February 0.78 0.13 5.93‡ 2.85 0.63 4.50‡ 8.96 1.74 5.14‡ March 0.70 0.15 4.67‡ 2.76 0.71 3.89‡ 9.00 1.94 4.65‡ April 0.38 0.20 1.91 1.91 0.99 1.92 5.55 2.60 2.13 May 0.06 0.25 0.22 0.20 1.32 0.15 1.44 3.38 0.43 June −0.16 0.29 −0.55 −0.51 1.57 −0.32 -0.63 3.97 -0.16 July −0.17 0.33 −0.54 0.12 1.77 0.07 1.06 4.46 0.24 August −0.28 0.32 −0.89 −0.16 1.73 −0.09 0.35 4.37 0.08 September −0.46 0.28 −1.66 -0.90 1.50 -0.60 -2.30 3.80 -0.61 October −0.42 0.21 −1.98 −0.82 1.12 −0.73 −2.80 2.88 −0.97 November −0.32 0.12 −2.63 −0.98 0.66 −1.47 −3.54 1.69 −2.09 AR1 0.56 0.12 4.62‡ 0.52 0.14 3.71‡ 0.46 0.13 3.44‡ MA1 0.43 0.13 3.24† 0.27 0.16 1.73 0.43 0.13 3.22‡

Akaike info criterion (AIC) −0.13 3.12 5.05

Schwarz criterion (SC) 0.41 3.66 5.59

R2 0.84 0.77 0.83

_p_{< 0.05,}†_p_{< 0.01,}‡_p_{< 0.001}

Selection of the final parameters was based upon the lowest AIC and SC. Reference group: December. Abbreviations: AR1= autoregressive, lag 1; MA1 = moving average, lag 1.

(11)

would have resulted in a 5% reduction of the number of deaths observed, with a vaccine effectiveness of 40%, whereas a 13% reduction could be expected for a vaccine effectiveness of 80% (Carrat & Valleron, 1995; Nguyen-Van-Tam et al., 2001).

The strength of our study lies in its longitudinal base and large popu-lation size, combined with the use of a comprehensive time-series analysis approach applied to the data of the age and sex subgroups. The method adopted was to fit the ARIMA models with the time series of the pneumo-nia admission rates; this is a method that has been widely used to examine the association between climate and the incidence of other dis-eases. The prior literature has noticeably documented the relationship between weather and pneumonia incidence rates using only univariate statistical analyses; however, given the high correlation between the meteorological parameters of each season, univariate analysis is not at all conducive to the identification of significant contributory factors. These results add significantly to our understanding of sex and age differences, as well as overall trends in pneumonia hospitalizations in a subtropical region of the Northern Hemisphere. Unlike prior studies that included participants from diverse ethnic groups, more than 98% of Taiwan's resi-dents are of Chinese Han ethnicity. While the homogenous population may exempt our study from potential confounding by race, it also limits its generalizeability to other ethnic groups.

Several other limitations of this study deserve consideration. First, we were unable to analyze behavioral factors, such as smoking and heavy drinking, which were not available in the databases used. Studies have found that behavioral factors are significantly associated with different age and sex-specific models (Diaz et al., 2007; Ewig et al., 1998; Pereira & Escuder, 1998; Woodhead et al., 1987). Second, the initial clinical symp-toms of pneumonia are similar to clinical sympsymp-toms of the common cold, which may mean patients delay seeking treatment for the disease. Our study cannot analyze the period between the onset of the clinical symp-toms of pneumonia and the start of treatment (i.e., hospitalization date) for this disease. Despite these limitations, we found that the monthly pneumonia admissions rate was significantly associated with seasonality and was higher at times of low ambient temperature. Public heath auth-orities should provide health education for people preparing them for high pneumonia risk seasons, and they should also provide vaccines to reduce risk of community-acquired pneumonia in winter.

ACKNOWLEDGMENTS

This study is based in part on data from the National Health Insurance Research Database provided by the Bureau of National Health Insurance, Department of Health and managed by National Health Research

(12)

Institutes. The interpretation and conclusions contained herein do not represent those of Bureau of National Health Insurance, Department of Health or National Health Research Institutes.

The authors have no conflict of interest to report.

REFERENCES

Almirall J, Bolibar I, Vidal J, Sauca G, Coll P, Niklasson B, Bartolome M, Balanzo X. (2000). Epidemiology of community-acquired pneumonia in adults: A population-based study. Eur Respir. J. 15:757–763.

Bartlett JG, Mundy LM. (1995). Community-acquired pneumonia. N. Engl. J. Med. 333:1618–1624. Bartlett JG, Dowell SF, Mandell LA, File TM Jr., Musher DM, Fine MJ. (2000). Practice guidelines

for the management of community-acquired pneumonia in adults. Infectious Diseases Society of America. Clin. Infect. Dis. 31:347–382.

Carrat F, Valleron AJ. (1995). Influenza mortality among the elderly in France, 1980–90: How many deaths may have been avoided through vaccination? J. Epidemiol. Community Health 49:419–425. Crighton EJ, Moineddin R, Mamdani M, Upshur RE. (2004). Influenza and pneumonia

hospitaliz-ations in Ontario: A time-series analysis. Epidemiol. Infect. 132:1167–1174.

Diaz A, Barria P, Niederman M, Restrepo MI, Dreyse J, Fuentes G, Couble B, Saldias F. (2007). Etiology of community-acquired pneumonia in hospitalized patients in Chile: The increasing prevalence of respiratory viruses among classic pathogens. Chest 131:779–787.

Donaldson GC, Keatinge WR. (2002). Excess winter mortality: Influenza or cold stress? Observational study. BMJ. 324:89–90.

Dowell SF, Whitney CG, Wright C, Rose CE Jr., Schuchat A. (2003). Seasonal patterns of invasive pneumococcal disease. Emerg. Infect. Dis. 9:573–579.

Ewig S, Ruiz M, Mensa J, Marcos MA, Martinez JA, Arancibia F, Niederman MS, Torres A. (1998). Severe community-acquired pneumonia. Assessment of severity criteria. Am. J. Respir. Crit. Care Med. 158:1102–1108.

Gutierrez F, Masia M, Mirete C, Soldan B, Rodriguez JC, Padilla S, Hernandez I, Royo G, Martin-Hidalgo A. (2006). The influence of age and gender on the population-based incidence of com-munity-acquired pneumonia caused by different microbial pathogens. J. Infect. 53:166–174. Jennings LC, Anderson TP, Beynon KA, Chua A, Laing RT, Werno AM, Young SA, Chambers ST,

Murdoch DR. (2008). Incidence and characteristics of viral community-acquired pneumonia in adults. Thorax 63:42–48.

Johnstone J, Majumdar SR, Fox JD, Marrie TJ. (2008). Viral infection in adults hospitalized with com-munity-acquired pneumonia: Prevalence, pathogens, and presentation. Chest 134:1141–1148. Jokinen C, Heiskanen L, Juvonen H, Kallinen S, Karkola K, Korppi M, Kurki S, Ronnberg PR,

Seppa A, Soimakallio S, Stén M, Tanska S, Tarkiainen A, Tukiainen H, Pyörälä K, Mäkelä PH. (1993). Incidence of community-acquired pneumonia in the population of four municipalities in eastern Finland. Am. J. Epidemiol. 137:977–988.

Kim PE, Musher DM, Glezen WP, Rodriguez-Barradas MC, Nahm WK, Wright CE. (1996). Association of invasive pneumococcal disease with season, atmospheric conditions, air pollution, and the isolation of respiratory viruses. Clin. Infect. Dis. 22:100–106.

Lauderdale TL, Chang FY, Ben RJ, Yin HC, Ni YH, Tsai JW, Cheng SH, Wang JT, Liu YC, Cheng YW, Chen ST, Fung CP, Chuang YC, Cheng HP, Lu DC, Liu CJ, Huang IW, Hung CL, Hsiao CF, Ho M. (2005). Etiology of community acquired pneumonia among adult patients requiring hospitalization in Taiwan. Respir. Med. 99:1079–1086.

Lee HC, Hu CJ, Chen CS, Lin HC. (2008). Seasonal variation in ischemic stroke incidence and associ-ation with climate: A six-year populassoci-ation-based study. Chronobiol. Int. 25:938–949.

Lieberman D, Lieberman D, Porath A. (1996a). Seasonal variation in community-acquired pneumo-nia. Eur. Respir. J. 9:2630–2634.

Lieberman D, Schlaeffer F, Boldur I, Lieberman D, Horowitz S, Friedman MG, Leiononen M, Horovitz O, Manor E, Porath A. (1996b). Multiple pathogens in adult patients admitted with community-acquired pneumonia: A one year prospective study of 346 consecutive patients. Thorax 51:179–184.

(13)

Lieberman D, Lieberman D, Friger MD. (1999). Seasonal variation in hospital admissions for commu-nity-acquired pneumonia: A 5-year study. J. Infect. 39:134–140.

Lin HC, Lin SY, Lee HC, Hu CJ, Choy CS. (2008). Weekly pattern of stroke onset in an Asian country: A nationwide population-based study. Chronobiol Int. 25:788–799.

Lowen AC, Steel J, Mubareka S, Palese P. (2008). High temperature (30°C) blocks aerosol but not contact transmission of influenza virus. J. Virol. 82:5650–5652.

Macfarlane JT, Colville A, Guion A, Macfarlane RM, Rose DH. (1993). Prospective study of aetiology and outcome of adult lower-respiratory-tract infections in the community. Lancet 341:511–514. Mandell LA. (1995). Community-acquired pneumonia: Etiology, epidemiology, and treatment. Chest

108:35S–42S.

Mandell LA, Bartlett JG, Dowell SF, File TM Jr., Musher DM, Whitney C. (2003). Update of practice guidelines for the management of community-acquired pneumonia in immunocompetent adults. Clin. Infect. Dis. 37:1405–1433.

Marrie TJ, Durant H, Yates L. (1989). Community-acquired pneumonia requiring hospitalization: Five-year prospective study. Rev. Infect. Dis. 11:586–599.

Marston BJ, Plouffe JF, File TM Jr., Hackman BA, Salstrom SJ, Lipman HB, Kolczak MS, Breiman RF. (1997). Incidence of community-acquired pneumonia requiring hospitalization. Results of a population-based active surveillance study in Ohio. The Community-Based Pneumonia Incidence Study Group. Arch. Intern. Med. 157:1709–1718.

Moine P, Vercken JB, Chevret S, Chastang C, Gajdos P. (1994). Severe community-acquired pneumo-nia: Etiology, epidemiology, and prognosis factors. French Study Group for Community-Acquired Pneumonia in the Intensive Care Unit. Chest 105:1487–1495.

Muller-Pebody B, Edmunds WJ, Zambon MC, Gay NJ, Crowcroft NS. (2002). Contribution of RSV to bronchiolitis and pneumonia-associated hospitalizations in English children, April 1995–March 1998. Epidemiol. Infect. 129:99–106.

Nguyen-Van-Tam JS, Brockway CR, Pearson JC, Hayward AC, Fleming DM. (2001). Excess hospital admissions for pneumonia and influenza in persons ≥65 years associated with influenza epi-demics in three English health districts: 1987–95. Epidemiol. Infect. 126:71–79.

Oosterheert JJ, Bonten MJ, Hak E, Schneider MM, Hoepelman AI. (2003). Severe community-acquired pneumonia: What's in a name? Curr. Opin. Infect. Dis. 16:153–159.

Ortqvist A, Hedlund J, Grillner L, Jalonen E, Kallings I, Leinonen M, Kalin M. (1990). Aetiology, outcome and prognostic factors in community-acquired pneumonia requiring hospitalization. Eur. Respir. J. 3:1105–1113.

Pereira JC, Escuder MM. (1998). The importance of clinical symptoms and signs in the diagnosis of community-acquired pneumonia. J. Trop. Pediatr. 44:18–24.

Portaluppi F, Touitou Y, Smolensky MH. (2008). Ethical and methodological standards for laboratory and medical biological rhythm research. Chronobiol. Int. 25:999–1016.

Santos de Unamuno C, Llorente San Martin MA, Carandell Jager E, Gutierrez Garcia M, Riera Jaume M, Ramirez Rosales A, Pareja Bezares A, Corrales Nadal A. (1998). Site of care provision, etiology and treatment of community-acquired pneumonia in Palma de Mallorca. Med. Clin. (Barc.) 110:290–294 [in Spanish].

Saynajakangas P, Keistinen T, Tuuponen T. (2001). Seasonal fluctuations in hospitalisation for pneu-monia in Finland. Int. J. Circumpolar Health 60:34–40.

Shahid MA, Abubakar M, Hameed S, Hassan S. (2009). Avian influenza virus (H5N1): Effects of physico-chemical factors on its survival. Virol. J. 6:38.

Taiwan Department of Health (2009). Health Statistics in Taiwan 2008. Taiwan Department of Health.

Tsai CL, Brenner BE, Camargo CA Jr. (2007). Circadian-rhythm differences among emergency department patients with chronic obstructive pulmonary disease exacerbation. Chronobiol Int. 24:699–713.

Viegi G, Pistelli R, Cazzola M, Falcone F, Cerveri I, Rossi A, Ugo Di Maria G. (2006). Epidemiological survey on incidence and treatment of community acquired pneumonia in Italy. Respir. Med. 100:46–55.

Woodhead MA, Macfarlane JT, McCracken JS, Rose DH, Finch RG. (1987). Prospective study of the aetiology and outcome of pneumonia in the community. Lancet 1:671–674.