Economic burden of community-acquired pneumonia among elderly patients: a Japanese perspective
Konomura et al. Pneumonia
Economic burden of community-acquired pneumonia among elderly patients: a Japanese perspective
Keiko Konomura 0
Hideaki Nagai 1
Manabu Akazawa 0
0 Public Health and Epidemiology, Meiji Pharmaceutical University , 2-522-1, Noshio, Kiyose, Tokyo 204-8588 , Japan
1 Center for Pulmonary Diseases, National Hospital Organization Tokyo National Hospital , 3-1-1 Takeoka, Kiyose-shi, Tokyo 204-8585 , Japan
Background: This study aimed to estimate the economic burden of community-acquired pneumonia (CAP) among elderly patients in Japan. In addition, the study evaluated the relationship between total treatment cost and CAP risk factors. Methods: An administrative database was searched for elderly patients (≥ 65 years old) who had pneumonia (ICD-10 code: J12-J18) and an antibiotic prescription between 1 June 2014 and 31 May 2015. The all-cause total healthcare costs of outpatient and inpatient CAP episodes were calculated. Results: This study evaluated data from 29,619 patients with CAP who experienced 14,450 outpatient CAP episodes and/or 20,314 inpatient CAP episodes. The mean ages were 77.5 ± 8.0 years and 81.5 ± 8.2 years among the outpatient and inpatient groups, respectively. The median treatment costs were US$346 (interquartile range: $195-551) per outpatient episode and US$4851 (interquartile range: $3313-7669) per inpatient episode. More severe cases had increased treatment costs at the treating hospitals. Male sex, diabetes, chronic obstructive pulmonary disease, and liver dysfunction were associated with increased total treatment costs, while dementia, dialysis, and rheumatism were associated with high costs of treating a CAP episode. Conclusions: The economic burden of CAP might be decreased by reducing the number of hospitalizations for mild CAP and the incidence of severe CAP. Therefore, preventative care (e.g. oral hygiene or pneumococcus vaccination) is recommended for patients with related risk factors, such as male sex, older age, diabetes, chronic obstructive pulmonary disease, liver dysfunction, rheumatism, dementia, or dialysis.
Community-acquired pneumonia; Economic burden of disease; Cost analysis; A-DROP system; Invasive pneumococcal disease; Pneumonia
Community-acquired pneumonia (CAP) is a common
acute infectious disease among elderly people, and is
associated with high rates of hospital admission and mortality.
In Japan, there are approximately 1.9 million new CAP
cases every year, with approximately 70% of cases involving
patients who are >65 years old and approximately 70%
of these elderly patients being hospitalized [
pneumonia is the third leading cause of death among
elderly people in Japan. Age and male sex are known
risk factors for CAP among adults [
], and many
comorbidities are also risk factors for CAP, such as
chronic obstructive pulmonary disease (COPD), diabetes,
cancer, dementia, congestive heart failure, and liver
function failure [
]. Prescriptions for inhaled medication
or oral corticosteroids are associated with an increased
risk of CAP . Several methods are used to classify the
severity of CAP, and the most commonly used methods
for selecting the CAP treatment location are the CURB-65
and pneumonia severity index (PSI) systems. Cases with
higher PSI classification (most mild: class I, most severe:
class V) have increased values for mortality rate, length of
stay, rate of subsequent hospitalization, and rate of
admission to the intensive care unit (ICU) [
CAP is associated with both clinical and economic
burdens, based on its high incidence, admission rate, and
mortality rate [
]. Many studies from various countries
have estimated the economic burden of CAP among elderly
], and those studies have revealed that
increased treatment costs are associated with older age and
treatment setting . For example, Dutch patients who
were ≥50 years old had treatment costs of > US$5000
per CAP episode, and a Spanish population-based study
revealed that direct costs for outpatient and inpatient
CAP treatment were approximately US$200 and US$1700,
]. The total CAP treatment costs among
inpatients increase for PSI classes I–III and reach a plateau
for classes IV–V [
]. Sato et al. have also stratified the
CAP-related risk based on the patient’s immune status and
chronic comorbidities [
], and reported that the all-cause
total healthcare costs were higher for high-risk cases,
compared to low-risk cases. Although Japan is the most
aged country in the world, the economic burden of CAP
according to risk factors and severity remains unclear.
Furthermore, it is difficult to directly compare the treatment
costs in various countries because of differences in the
treatment approach and insurance systems. Nevertheless,
it would be useful for healthcare providers and policy
makers to understand the effects of risk factors and
severity on treatment costs, which could help facilitate
the appropriate distribution of medical resources.
There are several prophylactic treatments for CAP.
Pneumococcal disease among adults can be prevented
using the 13-valent pneumococcal conjugate vaccine or the
23-valent pneumococcal polysaccharide vaccine (PPV23).
Many countries, including Japan, are also introducing
agebased pneumococcal vaccination programs, and the PPV23
vaccine is thought to prevent 50–85% of invasive
pneumococcal disease (IPD) cases [
], although the vaccination
coverage rate remains low [
]. Oral hygiene is also
known to reduce pneumonia onset among elderly patients
. Thus, based on the aging global population, a strategy
is needed to reduce the economic burden of CAP by
identifying patients with high predicted healthcare costs
and targeting individuals who should receive prophylactic
treatment. However, the effects of CAP severity and risk
factors on treatment costs remain unclear, and these
factors are an important part of cost-effective or cost-utility
analyses. The present study used a large administrative
database to estimate the economic burden of CAP among
elderly Japanese patients, as well as the per-episode and
total treatment costs according to disease severity and risk
The retrospective protocol of this study was approved by
the ethical committee of Meiji Pharmaceutical University
(Tokyo, Japan). Patient and treatment records from 1 April
2010 to 31 May 2015 were obtained from an
administrative database that is maintained by Medical Data Vision
Co. Ltd. (Tokyo, Japan). Pneumonia episodes were
identified between 1 June 2014 and 31 May 2015, while patient
characteristics were searched up until 1 June 2014. The
database includes claims data from approximately 5
million patients who received treatment at 200 acute
care hospitals, and were tracked using the Japanese
Diagnosis Procedure Combination/Per-Diem Payment
System (DPC/PDPS). These mainly consisted of small- and
medium-sized hospitals that were distributed throughout
Japan. Previous reports have provided detailed
explanations of the DPC/PDPS [
], which is a case-mix
system that tracks patients based on their diagnoses and
procedures. Fixed medical payments are determined
according to the case-mix system and additional fees
for service. In addition to the medical costs that are
calculated using the DPC/PDPS, the database also contains
information regarding the total healthcare costs based on
the provided medical services and their related fees.
Anonymized data in the database include demographic
characteristics, prescriptions, diagnosis (based on the
International Classification of Diseases, 10th revision
[ICD-10] code), medical treatments, treatment dates,
hospital admission and discharging dates, discharge
status, and pneumonia severity scores at the admission
based on the A-DROP system, which is advocated by
the Japanese Respiratory Society [
]. The A-DROP
system uses a modification of the CURB-65 score,
which has been adjusted to reflect Japan’s aging population.
The system evaluates five factors: age of >69 years for male
patients and >74 years for female patients, blood urea
nitrogen of >20 mg/dL or dehydration, an SpO2 value of
≤90% or a PaO2 value of ≤60 Torr, presence of
disorientation, and a systolic blood pressure of ≤90 mmHg. Cases
are defined as mild, moderate, severe, or very severe, based
on total scores of 0, 1–2, 3, and 4–5, respectively. The
database also includes outpatient treatment records that
are not covered by the DPC/PDPS.
Definition of CAP episodes
The present study included patients who were ≥65 years
old with a confirmed diagnosis of pneumonia (ICD-10
code: J12–J18) and a prescription for antibiotic treatment
between 1 June 2014 and 31 May 2015. The CAP episodes
were classified as outpatient episodes or inpatient
episodes. The outpatient index date was defined as the
date of a pneumonia diagnosis with any antibiotic
prescription. The treatment period extended from the
index date until the end of the antibiotic prescription.
If a second antibiotic prescription was provided within
7 days after the previous prescription’s end date, the
second treatment was considered part of the same
outpatient episode. The inpatient index date was defined
as the admission date for cases with a diagnosis of
pneumonia, and the treatment period was defined as the length
of stay (LOS). Inpatient CAP episodes were categorized
based on treatment in the general ward or ICU (a
minimum 1-night stay in the ICU). Death records (all-cause
deaths based on discharge records) were only available for
inpatient episodes. To ensure the study only considered
cases of CAP, patients with a discharge date that was
≤14 days before the index date were excluded.
CAP-related costs can be calculated as the CAP-related
healthcare costs and/or the all-cause total healthcare
costs. The CAP-related healthcare costs are calculated
based on treatments that are directly related to the CAP,
while the all-cause total healthcare costs are calculated
based on the total costs during a CAP treatment period.
The former method is difficult to use for elderly patients
with CAP, as they typically have one or more
comorbidities, which makes it difficult to differentiate between the
treatments for the CAP and the exacerbation of
comorbidities. Thus, the present study used the all-cause total
healthcare costs in the analyses.
All-cause total healthcare costs were calculated by
combining all recorded treatment costs during a single
CAP episode. The treatment costs were classified as
being related to inpatient stays, office visits, drug
treatments, examinations, medical procedures, and other
costs. The costs for inpatient stays and office visits
included the facility’s fixed costs and any service fees.
The costs for drug treatments, examinations, and other
costs were variable. The costs for medical procedures and
other categories included service fees. A detailed
breakdown of these categories is shown in Additional file 1:
Table S1. Treatment costs were estimated based on
medical fees and drug prices from April 2014. Because the
fees and prices are revised every 2 years in Japan, the
corresponding values for the period during which the
treatment costs were calculated were selected.
The A-DROP system scores range from 0 to 5, with the
most severe cases assigned a score of 5. CAP severity
was also defined based on IPD episodes (severe CAP), as
the related care is typically provided in a hospital. The
IPD diagnoses were classified as bacteremia or meningitis.
Bacteremia cases were identified based on a bacteremia
diagnosis (ICD-10: A403, A409, A419, A491, A499), blood
culture records, and antibiotic drug treatment during the
CAP treatment period. Meningitis cases were identified
based on a meningitis diagnosis (ICD-10: G001, G009),
lumbar puncture, and antibiotic drug treatment during
the CAP treatment period.
Comorbidities were defined based on the patients’ previous
medications and/or diagnoses before the index date. The
present study considered diabetes (ICD-10: E11–E14),
COPD (ICD-10: J42–J44), dementia (ICD-10: F00), liver
function failure (ICD-10: K70–K76), rheumatism (ICD-10:
M059, M060, M068, M069), and cancer (ICD-10: C00–
C99, D00–D09). Liver function failure does not have
any specific medications, and was defined for the study
as the presence of ≥2 diagnoses before the index date.
Severe renal dysfunction was identified based on at
least one dialysis treatment before the index date. The
detailed definitions are shown in Additional file 1:
Table S2. Medications that affected pneumonia were
defined as oral steroids, inhaled steroids,
angiotensinconverting enzyme inhibitors, and statins. To be
considered as influencing the development of CAP, these drugs
had to have been prescribed at least once during the
6 months before the index date. The definitions of
medications are listed in Additional file 1: Table S3.
All-cause total healthcare costs were calculated for
outpatient and inpatient CAP episodes, as well as breakdowns
of the total treatment costs. The costs of inpatient CAP
episodes were also subdivided into cases that received
treatment in the general ward or in the ICU. Treatment
costs were also calculated according to mortality and CAP
severity (based on the A-DROP score and episodes with
or without IPD). The costs were stratified according to the
presence of comorbidities and prescriptions, and all costs
were expressed as median and interquartile range (IQR)
values in US dollars (in May 2017, one US dollar equaled
approximately 111.3 yen). All results were reported as
mean, standard deviation, median, and IQR, as
appropriate. All analyses were performed using SAS software
(version 9.3; SAS Institute Inc., Cary, North Carolina, USA).
There were 70,539 patients who were ≥65 years old and
had a pneumonia diagnosis in the database. The study
evaluated data from 29,619 patients with CAP, who
experienced 14,450 outpatient CAP episodes and/or 20,314
inpatient CAP episodes. The study selection criteria is
shown in Fig. 1. The mean ages were 77.5 ± 8.0 years and
81.5 ± 8.2 years among the outpatient and inpatient groups,
respectively. Across both treatment settings, 61% of the
patients were men (Table 1). The median CAP treatment
periods were 7 days (IQR: 4–9 days) per outpatient episode
and 14 days (IQR: 9–25 days) per inpatient episode. Liver
function failure was the most prevalent comorbidity among
both inpatient and outpatient cases, and was followed
by COPD. Outpatient episodes were associated with a
higher rate of cancer, compared to inpatient episodes
(16% vs. 7%).
All-cause CAP treatment costs
The median treatment costs were $346 (IQR: $195–551)
per outpatient episode and $4851 (IQR: $3313–7669)
per inpatient episode. Drug costs and laboratory test
costs accounted for 82% of the outpatient CAP episode
costs, while 61% of inpatient treatment costs were
related to the inpatient stays (Fig. 2). Approximately 58%
of the episodes (20,145/34,764) involved hospitalization
in the general ward, and relatively few episodes
involved ICU treatment (0.5%, 169/34,764). However,
the treatment costs of ICU episodes were
approximately 2.6× higher, compared to general ward episodes
($12,728 [IQR: $8059–21,512] vs. $4824 [IQR: $3301–
7591], respectively). The combined total cost of CAP
during the study period was $136,575,963, with outpatient
episodes accounting for 6% of these costs, general ward
episodes accounting for 92%, and ICU episodes
accounting for 2%. The median treatment costs for cases
of in-hospital mortality (n = 2389) were significantly
higher than cases with survival until discharge ($6474
[IQR: $3372–10,639] vs. $4741 [IQR: $3308–7287],
CAP severity data were available for 83% of the inpatient
episodes (16,931/20,314). The mean treatment costs of
CAP episodes according to severity score are shown in
Table 2. Higher severity was associated with increased
treatment costs at the treating hospital, as well as with a
higher mortality rate and prolonged treatment period.
Approximately 1.3% of inpatient episodes involved IPD
(263/20,314), and 4.6% (12/263) of the IPD episodes were
treated in the ICU. The mean age of patients with IPD
episodes was 80.7 ± 7.5 years, and 67% of the patients were
men. The LOS for IPD episodes was longer than the LOS
for non-IPD episodes (21 days vs. 14 days). The mortality
rate for non-IPD episodes was 11%, compared to 35%
for IPD episodes. Additional detailed data are shown in
Additional file 1: Table S4. The median treatment cost
per IPD episode was $7766 (IQR: $4411–12,762), and
that cost was higher than the cost for non-IPD episodes
($4831 [IQR: $3304–7603]). Approximately 62% of the
IPD episodes (162/263) had available CAP severity data.
Cases with a severity score of 5 had the highest IPD
rate (8%). Figure 3 shows the median treatment costs
and mortality rates for IPD and non-IPD episodes. IPD
episodes had higher median treatment costs and an
increased mortality rate. Compared to non-IPD
episodes, the mortality rate for IPD episodes was increased
Outpatient CAP episodes
n = 14,450
Inpatient CAP episodes
n = 20,314
All-cause CAP treatment costs according to risk factors
Total costs were associated with male sex, diabetes,
COPD, and liver function failure (Table 3). Approximately
90% of the total costs were spent on general ward episodes
for each risk factor. Increasing age was associated with
higher hospitalization rates and total treatment costs.
CAP episodes that involved dementia and dialysis were
also associated with high admission rates. Table 4 shows
the median treatment costs per CAP episode, the median
treatment periods, and the mortality rates according to
the risk factors. Inpatient CAP episodes that involved
dementia, dialysis, and rheumatism were associated with
high treatment costs (likely because of the prolonged
LOS) and high mortality rates. Inpatient CAP episodes
that involved cancer were associated with a high mortality
rate, a short LOS, and low treatment costs.
The present study evaluated the economic burden of CAP
in Japan according to severity, as well as the relationship
between treatment costs and risk factors. The study found
that CAP episodes with hospitalization were 14× more
expensive, compared to outpatient episodes, and that
stayrelated costs accounted for 61% of inpatient treatment
costs. It also found that CAP severity, male sex, diabetes,
COPD, and liver function failure were associated with
high total treatment costs. Furthermore, CAP episodes
that involved dementia, dialysis, and rheumatism were
associated with high treatment costs, which was likely
because of the high admission rate. However, CAP
episodes that involved cancer were associated with low
inpatient treatment costs, which was likely related to
the high mortality rate and short LOS.
Although it was observed that severe CAP, based on
the A-DROP system, was associated with high treatment
costs, no significant relationship was observed between
cost and CURB-65 in a prospective non-interventional
clinical study [
]. Another study revealed that total
CAP community-acquired pneumonia, GW general ward, ICU intensive care unit, COPD chronic obstructive pulmonary disease, ACE-I angiotensin-converting
CAP treatment costs among inpatients increased for PSI
classes I–III and reached a plateau at classes IV–V [
Thus, the relationship between CAP severity and
treatment costs may differ depending on the severity score
that is used.
The most important factor that influenced treatment
cost was hospitalization, which accounted for 94% of
total costs. Furthermore, treatment of episodes in the
general ward was associated with high total treatment
costs, while treatment of episodes in the ICU only
accounted for 2% of total costs, despite these episodes
clearly involving serious cases (i.e. high treatment costs,
high mortality rate, and prolonged LOS). Interestingly,
the costs of drug treatment and laboratory testing did
not account for a large proportion of the total inpatient
costs, and a prospective observational study in a Spanish
public tertiary hospital also confirmed that the costs
could be broken down into room costs (69%), drug costs
(13%), laboratory costs (12%), and diagnostic procedures
]. Therefore, the economic burden of CAP could
be reduced by decreasing the number of admissions.
A previous study revealed that 61% of mild CAP cases
were admitted to a hospital [
], although the rate was
only 6% in the present study. This is likely related to
differences in the definitions that are used for the
CURB-65 and A-DROP systems, as the A-DROP system
has one mild category (score 0) and the CURB-65 system
has two mild categories (scores 0 and 1). When the
CURB-65 system was applied, the results indicated that
34% of mild CAP cases were hospitalized, which suggests
that the A-DROP system may underestimate the
proportion of mild CAP cases. Thus, the present study’s results
highlight the importance of minimizing the hospitalization
of mild CAP cases, and other previous studies have also
provided similar findings [
In the present study, the cost per inpatient episode
tended to increase with age, although a Dutch
administrative database study revealed that the median treatment
cost per episode in the general ward remained relatively
stable at approximately $5600 for patients who were
≥50 years old [
]. This difference is likely related to the
fact that the LOS increased with age in the present study,
but was stable in the Dutch study. For Dutch outpatient
episodes, the cost was approximately $850 per episode for
patients who were 50–84 years old, while the costs were
approximately 50% lower for patients who were ≥85 years
GW general ward, ICU intensive care unit, COPD chronic obstructive pulmonary disease, ACE-I angiotensin-converting enzyme inhibitors
old. However, in the present study, the treatment cost per
CAP episode only decreased slightly for patients who were
≥85 years old. One possible explanation for this difference
is that Dutch patients who were ≥85 years old had shorter
outpatient treatments, as most patients in this age group
Based on the findings of the present study, prophylactic
treatment (e.g. oral hygiene or pneumococcal vaccination)
is recommended to reduce treatment costs for patients
who are male and/or have comorbidities (e.g. diabetes,
COPD, and liver function failure). In addition, patients with
dementia, dialysis, or rheumatism may be an important
group to target, given their high treatment costs per CAP
The present study determined the exact CAP treatment
costs using a nationally representative data set, while
previous studies were limited by only evaluating patients at a few
hospitals. In addition, the present study’s calculations were
not influenced by changing costs based on the patient’s
income or insurance status, as Japanese medical service fees
are determined based on a payment list that is approved by
the government. Furthermore, healthcare claims data that
are used for reimbursement can provide an accurate record
of the provided medical services.
The study calculated the all-cause total healthcare
costs according to CAP severity and risk factors. This is
because the treatment of underlying disease could not
be separated from the treatment of CAP exacerbation.
By comparing patients with or without CAP, further
study is planning to estimate the additional costs of
CAP. The treatment costs of CAP episodes that were
encountered by general practitioners were not estimated,
as the study’s database only included hospital records.
However, the authors speculate that outpatient treatment
costs of CAP episodes are comparable between hospitals
and clinics. This is because almost all residents of Japan
are covered by the national health insurance system,
which provides equal reimbursement for hospital-based
outpatient visits and clinic visits. The reimbursement
rates are pre-determined by the Japanese government, and
even patients who visit a large hospital without a referral
would only make an extra out-of-pocket payment of <
$72. However, the study’s database mainly included
smallor middle-sized hospitals, which suggests that patients
would incur similar treatment costs at a hospital or clinic
if they received the same treatment. Moreover, patients
can select any hospital or clinic throughout Japan,
regardless of the severity of their disease. Therefore, it is expected
that there would be little difference in the outpatient
treatment costs of CAP episodes at hospitals and clinics.
The present study also has several limitations. First,
CAP cases that were caused by Streptococcus pneumoniae
were not able to be identified, as culture results were not
available in the database. Therefore, it was difficult to
distinguish between specified or unspecified pneumonia.
As there is very little evidence regarding IPD-related
treatment costs, further studies are needed to more precisely
estimate the IPD treatment costs using culture results.
Second, compared to outpatients in the database, the
hospitalized patients included fewer immunocompromised
patients (e.g. patients with cancer and rheumatism). It is
possible that many patients with cancer and a poor general
status were included in the database. Another explanation
is that all elderly people (> 65 years old) can receive the
PPV23 vaccine in Japan, although the vaccination rate in
2014 was only 20% (based on vaccine shipment records),
which could indicate that high-risk patients had been
vaccinated and were less likely to develop CAP or other
serious conditions. Third, a claims database was used for
the present study, and this type of study is limited by
various issues. For example, there can be differences between
diagnostic and reimbursement records, and it would be
useful to combine data regarding diagnoses and specific
medications. However, there is no specific medication for
treating liver function failure, and the present study
arbitrarily defined these cases as patients who had ≥2
diagnoses before their index date. Therefore, these
diagnoses may have been overestimated.
This study found that severe CAP cases had high rates
of mortality and prolonged hospitalization, which were
associated with high treatment costs. In addition, many
hospitalizations that involved cases of mild CAP were
detected. Therefore, to reduce the economic burden of
CAP, clinicians should aim to decrease the number of
hospitalized mild CAP cases and to prevent the
development of severe CAP. For example, prophylactic treatment
(e.g. oral hygiene or vaccination) may be effective for elderly
patients who are male, ≥ 85 years old, or have comorbidities
(e.g. diabetes, COPD, liver function failure, rheumatism,
dementia, or dialysis). These findings can be used to
identify population(s) that should be targeted by the vaccine
program, based on their burden of diseases, and will help to
estimate the value of the vaccine program. Furthermore,
vaccination coverage remains low in Japan, and efforts
should be made to increase pneumococcus vaccination
Additional file 1: Table S1. Definition of the six categories for the
breakdown of total treatment costs. Table S2. Definition of comorbidities
and their medications. Table S3. Definition of drug use. Table S4. The
median treatment costs and treatment period, and death rate of invasive
pneumococcal disease (IPD). (PDF 586 kb)
CAP: Community-acquired pneumonia; COPD: Chronic obstructive
pulmonary disease; DPC/PDPS: Diagnosis Procedure Combination/Per-Diem
Payment System; ICD-10: International Classification of Diseases, 10th
revision; ICU: Intensive care unit; IPD: Invasive pneumococcal disease;
IQR: Interquartile range; LOS: Length of stay; PPV23: 23-Valent pneumococcal
polysaccharide vaccine; PSI: Pneumonia severity index; SD: Standard deviation
This research was supported by the Emerging/Re-emerging Infectious
Diseases Project, which is administered by the Japanese Agency for Medical
Research and Development (AMED).
Availability of data and materials
The database is commercially available from Medical Data Vision Co. Ltd.
MA and KK conceived and designed the research plan. KK analyzed and
interpreted the patient data, under the supervision of MA, and was a major
contributor in writing the manuscript. NH assisted with the clinical
interpretation of the data. All authors read and approved the final manuscript.
Ethics approval and consent to participate
This study’s retrospective protocol was approved by the ethical committee
of Meiji Pharmaceutical University.
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Submit your next manuscript to BioMed Central
and we will help you at every step:
1. Morimoto K , Suzuki M , Ishifuji T , et al. The burden and etiology of community-onset pneumonia in the aging Japanese population: a multicenter prospective study . PLoS One . 2015 ; 10 : 1 - 18 .
2. Farr BM , Woodhead MA , MacFarlane JT , et al. Risk factors for communityacquired pneumonia diagnosed by general practitioners in the community . Respir Med . 2000 ; 94 : 422 - 7 .
3. Baik I , Curhan GC , Rimm EB , et al. A prospective study of age and lifestyle factors in relation to community-acquired pneumonia in US men and women . Arch Intern Med . 2000 ; 160 : 3082 - 8 .
4. Jackson ML , Neuzil KM , Thompson WW , et al. The burden of communityacquired pneumonia in seniors: results of a population-based study . Clin Infect Dis . 2004 ; 39 : 1642 - 50 .
5. Vila Corcoles A , Ochoa Gondar O , Rodriguez Blanco T , et al. Epidemiology of community-acquired pneumonia in older adults: a population-based study . Respir Med . 2009 ; 103 : 309 - 16 .
6. Millett ERC , De Stavola BL , Quint JK , et al. Risk factors for hospital admission in the 28 days following a community-acquired pneumonia diagnosis in older adults, and their contribution to increasing hospitalisation rates over time: a cohort study . BMJ Open . 2015 ; 5 : e008737 .
7. Benjamin AL , Edward JB , Thomas SI , et al. Risk factors for acquiring pneumococcal infections . Arch Intern Med . 1986 ; 146 : 2179 - 85 .
8. Koivula I , Sten M , Makela PH . Risk factors for pneumonia in the elderly . Am J Med . 1994 ; 96 : 313 - 20 .
9. Almirall J , Bolíbar I , Balanzó X , et al. Risk factors for community-acquired pneumonia in adults: a population-based case-control study . Eur Respir J . 1999 ; 13 : 349 - 55 .
10. Sanz Herrero F , Blanquer OJ . Microbiology and risk factors for communityacquired pneumonia . Semin Respir Crit Care Med . 2012 ; 33 : 220 - 31 .
11. Fine MJ , Auble TE , Yealy DM , et al. A prediction rule to identify low-risk patients with community acquired pneumonia . N Engl J Med . 1997 ; 336 : 243 - 50 .
12. Sato R , Gomez Rey G , Nelson S , et al. Community-acquired pneumonia episode costs by age and risk in commercially insured US adults aged >/=50 years . Appl Health Econ Health Policy . 2013 ; 11 : 251 - 8 .
13. File TM Jr, Marrie TJ . Burden of community-acquired pneumonia in north American adults . Postgrad Med . 2010 ; 122 : 130 - 41 .
14. Huang SS , Johnson KM , Ray GT , et al. Healthcare utilization and cost of pneumococcal disease in the United States . Vaccine . 2011 ; 29 : 3398 - 412 .
15. Reyes S , Martinez R , Vallés JM , et al. Determinants of hospital costs in community-acquired pneumonia . Eur Respir J . 2008 ; 31 : 1061 - 7 .
16. Cascini S , Agabiti N , Incalzi RA , et al. Pneumonia burden in elderly patients: a classification algorithm using administrative data . BMC Infect Dis . 2013 ; 13 : 559 .
17. Tichopad A , Roberts C , Gembula I , et al. Clinical and economic burden of community-acquired pneumonia among adults in the Czech Republic, Hungary, Poland and Slovakia . PLoS One . 2013 ; 8 : e71375 .
18. Yoo KH , Yoo CG , Kim SK , et al. Economic burden and epidemiology of pneumonia in Korean adults aged over 50 years . J Korean Med Sci . 2013 ; 28 : 888 - 95 .
19. Thomas CP , Ryan M , Chapman JD , et al. Incidence and cost of pneumonia in medicare beneficiaries . Chest . 2012 ; 142 : 973 - 81 .
20. Kawakami K , Ohkusa Y , Kuroki R , et al. Effectiveness of pneumococcal polysaccharide vaccine against pneumonia and cost analysis for the elderly who receive seasonal influenza vaccine in Japan . Vaccine. 2010 ; 28 : 7063 - 9 .
21. Rozenbaum MH , Mangen MJ , Huijts SM , et al. Incidence, direct costs and duration of hospitalization of patients hospitalized with community acquired pneumonia: a nationwide retrospective claims database analysis . Vaccine . 2015 ; 33 : 3193 - 9 .
22. Bartolomé M , Almirrall J , Morera J , et al. A population-based study of the costs of care for community-acquired pneumonia . Eur Respir J . 2004 ; 23 : 610 - 6 .
23. Bauer TT , Welte T , Ernen C , et al. Cost analyses of community-acquired pneumonia from the hospital perspective . Chest . 2005 ; 128 : 2238 - 46 .
24. Butler JC , Breiman RF , Campbell JF , et al. Pneumococcal polysaccharide vaccine efficacy. An evaluation of current recommendations . JAMA . 1993 ; 270 : 1826 - 31 .
25. Poethko Müller C , Schmitz R . Vaccination coverage in German adults: Results of the German Health Interview and Examination Survey for Adults (DEGS1) . Vol. 56 . Bundesgesundheitsblatt - Gesundheitsforschung - Gesundheitsschutz; 2013 . p. 845 - 57 .
26. Australian Institute of Health and Welfare. 2009 Adult vaccination survey: summary results . 2011 . http://www.aihw.gov.au/publication-detail/? id= 10737418409. Accessed 15 Apr 2017 .
27. Yoneyama T , Yoshida M , Ohrui T , et al. Oral care reduces pneumonia in older patients in nursing homes . J Am Geriatr Soc . 2002 ; 50 : 430 - 3 .
28. Nakamura K. Diagnosis procedure combination database would develop Nationwide clinical research in Japan . Japanese Circ Soc . 2016 ; 80 : 2289 - 90 .
29. Milea D , Azmi S , Reginald P , et al. A review of accessibility of administrative healthcare databases in the Asia-Pacific region . J Mark Access Health Policy . 2015 . doi.10.3402/jmahp.v3. 28076 .
30. Tanaka S , Seto K , Kawakami K. Pharmacoepidemiology in Japan: medical databases and research achievements . J Pharm Heal Care Sci . 2015 . doi.10. 1186/s40780-015-0016-5.
31. Uematsu H , Kunisawa S , Yamashita K , et al. The impact of patient profiles and procedures on hospitalization costs through length of stay in community-acquired pneumonia patients based on a Japanese administrative database . PLoS One . 2015 ; 10 : 1 - 13 .
32. Shindo Y , Sato S , Maruyama E , et al. Comparison of severity scoring systems A-DROP and CURB-65 for community-acquired pneumonia . Respirology . 2008 ; 13 : 731 - 5 .
33. Miyashita N , Matsushima T , Oka M , et al. The JRS guidelines for the management of community-acquired pneumonia in adults: an update and new recommendations . Intern Med . 2006 ; 45 : 419 - 28 .
34. Cupurdija V , Lazic Z , Petrovic M , et al. Community-acquired pneumonia: economics of inpatient medical care vis-à-vis clinical severity . J Bras Pneumol . 2015 ; 41 : 48 - 57 .
35. Choudhury G , Chalmers JD , Mandal P , et al. Physician judgement is a crucial adjunct to pneumonia severity scores in low-risk patients . Eur Respir J . 2011 ; 38 : 643 - 8 .
36. Aliberti S , Ramirez J , Cosentini R , et al. Low CURB-65 is of limited value in deciding discharge of patients with community-acquired pneumonia . Respir Med . 2011 ; 105 : 1732 - 8 .
37. Spoorenberg SM , Bos WJ , Heijligenberg R , et al. Microbial aetiology, outcomes, and costs of hospitalisation for community-acquired pneumonia; an observational analysis . BMC Infect Dis . 2014 ; 14 : 335 .