2001 - Severity Assessment of Lower Respiratory Tract Infection in Elderly Patients in Primary Care

7/25/2019 2001 - Severity Assessment of Lower Respiratory Tract Infection in Elderly Patients in Primary Care

1/5

Severity Assessment of Lower Respiratory TractInfection in Elderly Patients in Primary Care

Yrjo Seppa, MD; Aini Bloigu, BSc; Pekka O. Honkanen, MD, PhD; Liisa Miettinen, MD; Hannu Syrjala, MD, PhD

Background: Simple markers for evaluating the sever-ity of lower respiratory tract infections (LRTI) in pri-mary care are lacking. It is of value to examine whetherthe information available to the primary care physicianduring a patients initial visit can be used to assess theseverity of LRTI.

Methods:The associations between different baselinevariables and outcomes (survival within or more than30 days) were investigated prospectively in a series of950 home-living patients 65 years or older with severe

LRTI that their primary care physicians suspected to bepneumonia.

Results:Twenty-one men and 17 women died (4.1%)within 30 days. According to univariate analysis, the fol-lowing parameters differed (P.01) between the fatali-ties and survivors: acute aggravation of a coexisting ill-

ness, age, respiratory rate, white blood cell count, andC-reactive protein (CRP) level. According to Cox for-ward stepwise regression analysis (P =.01 for entry and.05 for removal), acute aggravation of a concurrent ill-ness, respiratory rate (25/min), and CRP concentra-tion (100 mg/L) were independently associated withdeath.Themortality rate was 2.2% if thepatientshad noneor only 1 of the independent risk factors and 20% if theyhad all 3 risk factors.

Conclusions: Preceding aggravation of a concurrent ill-

ness and respiratory rate of 25/min or higher, togetherwith an elevated serum CRP level (100 mg/L), can beused as simple markers for identifying patients with thehighest risk for LRTI and improve management deci-sions among elderly people in primary care.

Arch Intern Med. 2001;161:2709-2713

L

OWER RESPIRATORYtract in-fections (LRTI) includeinfections of the tracheo-bronchial tree (bronchitis)and the lung parenchyma

(pneumonia). It has been estimated, basedon the data published from the UnitedKingdom, that a third of the adult pa-tients with LRTI in community settingsseek help for their symptoms. Antibioticagents are prescribed to a fourth of thesepatients, and community-acquired pneu-monia (CAP) is diagnosed in less than2% of the cases (less than 0.5% of thewhole adult population with LRTI incommunity settings).1 However, the inci-dence of CAP is higher than this in theelderly population.2,3

Community-acquired pneumonia is

the fifth most common reason for hospi-talization among patients aged 65 to 74years and the leading cause for hospital-ization among those 85 yearsor older.4 Thecost of treating CAP depends crucially onthe place of treatment.5 Safe strategies tominimize hospitalization would lead tomarked cost savings.6 Moreover, most pa-tients with CAP with little risk ofdeathpre-fer treatment outside the hospital,7 and,presently, most patients with CAP are ac-

tually treated as outpatients.1,8 Severalmodels have been developed for assess-ing the severity of CAP.9-18 However, thesemodels have been typically developedamong hospitalized patients withCAP and

are not necessarily applicable to outpa-tients with CAP or, more generally, to pa-tients with LRTI in primary care becausethe scoring systems in the models often re-quire both chest radiography and labora-tory analyses, which are not necessarilyavailable in primary care units.

The purpose of our prospective studywas to find out whether the informationavailable in theprimary care consulting of-fice can be used to assess the severity ofLRTI and the need for hospital treat-ment.Toward this end, we investigated theoutcomes of 950 home-living immuno-

competent elderly patients (age, 65years) with LRTI who had signs sugges-tive of pneumonia.

RESULTS

PATIENT POPULATIONAND MORTALITY

Thefinal study population consisted of 950home-living elderly patients with LRTI,38

ORIGINAL INVESTIGATION

From the Department ofInfection Control, OuluUniversity Hospital, Oulu

(Drs Seppa and Syrjala); theNational Public HealthInstitute, Oulu (Ms Bloigu); theDepartment of Public HealthScience and General Practice,University of Oulu, Oulu(Dr Honkanen); OuluUniversity Hospital, Oulu, andOulaskangas Hospital,Oulainen (Dr Miettinen); andKemi Health Center, Kemi(Dr Honkanen), Finland.

(REPRINTED) ARCH INTERN MED/VOL 161, DEC 10/24, 2001 WWW.ARCHINTERNMED.COM2709

2001 American Medical Association. All rights reserved.

wnloaded From: http://archinte.jamanetwork.com/ on 10/06/2013


2/5

(4.1%) of whom died within 30 days after the initial visit.Of these deaths, 37% occurred within the first 7 days(Figure). Their deaths were not dependent on the ini-tial place of treatment (Table 1). Themortality rates were

similar for both sexes (Table 2). In this population withLRTI, dyspnea at rest tended to be more often observedin the patients who died within 30 days (63%) than inthose who survived (48%) (P =.07; Table 2), while the

PATIENTS AND METHODS

PATIENTS

The study was carried out in association with a pneumo-coccaland influenza trial among persons 65 yearsor olderin northern Finland.19 During September and October 1992,before the baseline of the study, one of the researchers

(P.O.H.) visited all the municipal health centers and mu-nicipal district hospitals in thefollow-up area to inform thephysicians about the research design and the way of re-porting patients with LRTI. The follow-upvisits to the par-ticipating centers were made in the autumns of 1993 and1994. The study population consisted of all persons 65 yearsor older living in 55 municipalities in northern Finland.Thecensus on December 31,1992, included 59790 people,38.8% of whom were male. The study protocol was ap-proved by the ethical review committee of the medical fac-ulty of the University of Oulu, Oulu, Finland. Enrollmentin the follow-up study was voluntary. The physician madethe diagnosis of LRTI on the basis of the patients historyand physical examination findings. During the whole studyperiod, the primary care physicians made their decisions

concerning the prescription of antibiotics and the pos-sible referral of their patients independently.

METHODS

During the 2-year study period, 1743 cases of LRTI withsigns suggestive of pneumonia were reported among thestudy population of 59790 persons 65 years or older. Thenumber of patients visiting health centers with signs sug-gestive of pneumonia is not known, nor is the proportionof patients with LRTI butwithout signs suggestive of pneu-monia. Of the cases, 22 were excluded: 21 because of miss-ing information on the day of the initial visit and 1 be-cause of age (60 years). One episode of suspected LRTIper patient was reported in 1424 cases. Only the first epi-

sode was included in the study for 133 patients with morethan 1 episode of LRTI. Only home-living patients were in-cluded in the further analysis (1072 cases). Patients re-ported to be in a terminal state (10 cases) or with demen-tia (89 cases)were excluded because our questionnaire didnot include information aboutwhether their treatment wasactive or not. Of the remaining patients, those reported tobe bedridden (14 cases) were excluded because our pri-mary goal was to investigate an elderly population ca-pable of normal daily activities. Nine patients receiving im-munosuppressive medication (5 mg/d of cytostatics orprednisolone) were excluded.

The following patientinformation was recorded on thecase report form by the attending physician to describe thepatients condition during the first visit: presence or ab-sence of respiratory and other symptoms (cough, dyspnea

at rest, pleuritic chest pain, acute confusion, acute dete-rioration of general condition, and/or acute aggravation ofa coexisting chronic disease [eg, the impairment of glu-cose balance in diabetes or the deterioration of congestive

heart failure]); duration of symptoms; date of examina-tion; residential information (home, nursing home, mu-nicipal health center, or hospital ward); initial place of treat-ment (home, nursing home, municipal health centerward, or hospital ward); results of the initial physicalexamination (body temperature, respiratory rate, systolicanddiastolic blood pressure, heart rate, andabnormalfind-ing on chest auscultation, ie, rales); and whether the pa-tient wasin a terminal state. The patient data were comple-

mented during the follow-up visit (or after death, if thepatientdied before the scheduled follow-up visit) with thebasic laboratory data recorded at the initial visit: hemoglo-bin and C-reactive protein (CRP) levels; platelet and whiteblood cell (WBC) counts; erythrocyte sedimentationrate; concurrent illnesses or disabilities (eg, congestive heartfailure, asthma, chronic obstructive pulmonary disease,dementia, chronic pyelonephritis, and/or type of diabe-tes); residence on a long-term ward; bedridden state;immunosuppressive treatment (ie, 5 mg/d of cytostaticmedication or prednisolone); previous and current smok-ing habits; consumption of alcohol; information about pos-sible travel abroad within the past month; final diagnosis;and, in fatal cases, whether the death was due to LRTI. Toobtainvalid and accurate information on the date of death

andto confirm theregistered coexisting illnesses, data con-cerning the study population were also drawn from the na-tional register of the Finnish Social Insurance Institution,Kela. Information on dementia, dietary diabetes, alcoholabuse, and bedridden and terminal state was not availablein thenational register andwas therefore obtained by a ques-tionnaire. All data were stored in a computer database.

STATISTICAL ANALYSIS

The end point of the severity of LRTI was defined asmortality due to LRTI within 30 days after the first visit toa primary carephysician.16 The statistical analysis was per-formed using the SPSS software (SPSS Inc, Chicago, Ill).Survival after the initial visit wascalculatedwith the Kaplan-

Meier method. For the analysis of categorical variables, the2 test (or the Fisher 2-tailed exact test when appropriate)was used. The continuous variables were skewed, andmedian values and interquartile ranges (25th to 75th per-centile) were therefore used. The association between thecontinuous variables and survival was analyzed using theMann-Whitney test. For further analysis, the statisticallysignificant continuous variables were dichotomized by se-lecting clinically relevant cutoff points. To identify the in-dependent riskfactors among the variablesthat showed sta-tistically significant associations (P.01) with mortality inunivariate analyses, the relative risk of death was esti-mated using hazard ratios calculated by Cox forwardstep-wise regression analysis. The P value forentry into themodelwas .01 and for removal, .05. The patients having none oronly 1 missing value of predictor variables were included

for Cox regression analysis, and missing data were re-placed by the geometric means of the study population: 51mg/L for CRP level, 22/min for respiratory rate, and9.3103/L for WBC count.





3/5

presence of rales did not statistically differ between thegroups (81% vs 85%). Most of our patients with LRTIactually had real pneumonia; 83% of them had rales onchest auscultation, and 48% had dyspnea at rest duringthe first visit to a primary care physician. Moreover, re-sults of a retrospective analysis of the chest roentgeno-grams (CRXs) showed definite pneumonia in 61% andprobable pneumonia in 13%. However, we excluded theCRX findings from this study to mimic the real condi-tions because CRX was only available in 44% of the casesat the time of the first visit to a primary care physicianin our series.

SIGNIFICANT VARIABLES ASSOCIATEDWITH MORTALITY

Of the categorical variables, only acute aggravation of acoexisting illness had a statistically significant associa-tion with survival (Table 2). However, the comorbiditywas not associated with mortality in univariate analysis(data not shown). Of the continuous variables, the pa-tients age, respiratory rate, WBC count, and CRP val-ues were statistically associated with death within 30 days(Table 3). When the importance of these variables wasevaluated in the Cox forward stepwise regression modelthat included 719 cases, acute aggravation of a coexist-ing illness, respiratory rate (25/min), and CRP level(100 mg/L) were identified (in this order of magni-

tude of hazard ratios) as independent relative risk fac-tors of death within 30 days (Table 4). The mortalityrate of these patients with LRTI was 2.2% within 30 daysif they had none or only 1 independent risk factor. Thecorresponding mortality rate was 20% if they had all 3risk factors.

COMMENT

In this study, 2 parameters immediately available in theconsulting office (preceding aggravation of a coexistingillness and respiratory rate of25/min) together withan elevated serum CRP level (100 mg/L) at the initialvisit to a primary care unit, were independently associ-ated with the risk of death within 30 days among elderlypatients with LRTI and can be used to assess the sever-ity of LRTI.

As our results show, in most of thepatients with LRTIin whom the attending physicians had primarily sus-pected to have pneumonia, the diagnosis of CAP waslaterverified by CXR. Thus, a comparison of our results withearlier reports concerning mainly CAP seemed justifiable.

It is well known that elderly patients may lack thetypical symptoms of pneumonia; the earliest clues maybe unspecific symptoms, such as lethargy, mental con-fusion, andfailure to thrive,as well as the deteriora-tionof a preexistingdisease(eg, congestiveheart failure).20

In our series, acute aggravation of a coexisting illness,such as the impairment of glucose balance in diabetes orthe deterioration of congestive heart failure, was inde-pendently associated with mortality within 30 days. This

100

99

98

97

96

950 5 10 15 20 25 3530

Time, d

CumulativeSurvival,%

Survival after the initial visit among 950 home-living elderly patients withlower respiratory tract infection in primary care during 30 days (note thescale on the y-axis).

Table 1. Initial Place of Treatment and MortalityWithin 30 Days Among Home-Living Elderly PatientsWith Lower Respiratory Tract Infections*

Initial Placeof Treatment

No. of Deaths/No. of Patients Mortal ity, %

Home 10/257 3.9

Nursing home 0/4 0.0

Health center ward 19/420 4.5

Hospital ward 9/256 3.5

Total 38/937 4.1

*P.2, Fisher 2-tailed exact test.Information was not available for 13 patients.

Table 2. Essential Clinical Data of Home-Living Elderly Patients Seeking CareFrom Primary Care Physicians for Lower Respiratory Tract Infection

Variable

Death Within 30 d Survival 30 d

P Value*

No. of Observations Positive, % No. of Observations Positive, %Male sex 38 55.3 912 51.8 .2

Cough 38 68.4 896 77.5 .2

Dyspnea at rest 38 63.2 895 47.9 .07

Pleuritic pain 37 35.1 876 37.7 .2

Acute confusion 38 2.6 868 6.3 .2

Acute aggravation of

General condition 37 86.5 892 73.1 .07

Coexisting illness 38 39.5 875 20.9 .007

Abnormal chest auscultation (rales) 36 80.6 888 85.0 .2

*By 2 test (Fisher 2-tailed exact test when appropriate).





4/5

acute aggravation of a chronic illness was judged by thecriteria of the attending physicians.

The measurement of respiratory rate is a usefulmarker of acute respiratory dysfunction.21 Its prognos-tic importance in CAP has been shown in several in-vestigations.9-12,14-16,18 The cutoff value of respiratoryrate has varied from 20/min or higher22 to 30/min orhigher.9-12,14-16,18 In our study, tachypnea (with respira-tory rates of25/min used as a cutoff point) remained asignificant predictor of mortality even in multivariateanalysis. Our results stress once again the importance ofrespiratory ratea simple measurementin the clini-

cal assessment of respiratory infections. On theother hand,heart rate, systolic and diastolic blood pressure, and bodytemperature, which havepreviously been reportedas im-portant risk factors associated with death in CAP, did nothave similar predictive value in the present series.

Because CRP level is more sensitive, decreases fasterafter a favorable treatment response, and is independentof age, its use as a marker instead of erythrocyte sedi-mentation rate has been recommended for the diagno-sis and follow-up of infections.23,24 In adults with acuteLRTI, a CRP level of 50 to 75 mg/L has been considered

suggestive of CAP.1,23 Our results showed that a CRP level100 mg/L or higher was independently associated with

mortality in elderly patients with LRTI in primary care.Thus, our study emphasizes the importance of CRP inassessing the severity of LRTI among elderly patients inprimary care. Our results also suggest that the use of CRPas a marker is recommendable in the severity assess-ment of respiratory tract infections.

In our series, the overall rate of mortality of elderlyhome-living patients with LRTI within 30 days was 4%,which was lower than the percentage published in, forexample, a recent meta-analysis in which the mortalityrate of elderly hospitalized patients with CAP varied from5.7% to 32.9%.22 In our primary care population, the mor-tality rate was even lower (2%) if the patients had no oronly 1 of theindependent risk factors associated with mor-

tality. However, the mortality rate was 20% among ourpatientswith 3 positive risk factors. The latter figure agreeswith an earlier report on mortality (21%) from pneumo-coccal bacteremia in Finland.25

In earlier studies, controversial observations on theinfluence of age on the mortality in CAP have been re-ported, with some studies supporting its impor-tance11,17,26,27 and others not.28-30 In our series, age was asignificant factor in univariate analysis but did not re-main significantly associated with mortality in multivar-iate analysis. In many series, comorbidity has been shownto be an independent risk factor of mortality.16,17,22 Nosuch association was seen in our series. In the earlier stud-ies in which comorbidity has been recognized as a risk

factor, patients with CAP have been typically treated inthe hospital. In earlier studies, leukopenia and leukocy-tosis have also been shown to be independent prognos-tic markers in CAP.9,11,18,22,27 In our series, WBC countwas significantly associated withdeath in univariate analy-sis, but when contrasting the patients with a WBC countof less than 4103/L or higher than 11103/L withthose with a WBC count within the normal range, WBCcount wasno longer independently associated with deathin Cox regression analysis. Whether the use of cytostat-ics or steroids could be an independent prognostic vari-

Table 3. Clinical and Laboratory Findings of Home-Living Elderly Patients With Symptomatic Lower Respiratory Tract Infection

Variable

Death Within 30 d Survival 30 d

P ValueNo. of Cases Median IQR* No. of Cases Median IQR*

Age, y 38 80 73-86 912 76 71-81 .003

Duration of symptoms, d 28 3 1-5 843 3 1-7 .2

Physical findings on admission

Body temperature,C 35 37.5 37.0-38.3 871 37.6 36.8-38.4 .2

Respiratory rate, per minute 31 26 21-36 675 22 18-27 .002

Heart rate, per minute 37 89 70-104 832 80 72-94 .2Systolic blood pressure, mm Hg 36 135 116-164 859 142 130-160 .2

Diastolic blood pressure, mm Hg 36 79 70-89 855 80 70-90 .2

Laboratory findings on admission

Hemoglobin, g/dL 29 13.0 11.6-13.5 627 13.3 12.1-14.3 .08

Platelets, 103/L 18 230 155-371 320 252 191-320 .2

Leukocytes, 103/L 29 12.2 9.4-18.3 667 9.3 7.0-12.2 .001

Erythrocyte sedimentation rate, mm/h 15 52 40-84 515 38 20-66 .04

C-reactive protein, mg/L 27 136 80-214 597 76 24-142 .001

*Interquartile range (25th to 75th percentile).Mann-Whitney test.

Table 4. Cox Forward Stepwise Regression Analysisof Independent Risk Factors

Variable

Relative Riskof Death

(95% CI*)P Value

to Remove

Acute aggravation of coexisting illness .006

No 1.0

Yes 2.8 (1.4-5.6)

Respiratory rate, per minute .005

25 1.0

25 2.8 (1.4-5.5)

C-reactive protein, mg/L .01

100 1.0

100 2.5 (1.2-5.1)

*CI indicates confidence interval.





5/5

able even in primary care patients should be investi-gated in a population with a larger proportion of patientsundergoing such treatments.

Recently, a prognostic model for the evaluation ofthe severity of CAP has been described15 and validatedin elderly hospitalized patients with CAP.31We were un-able to use this prognostic system, however, as some ofthe laboratory measurements essential for the calcula-tion of the risk scores of individual patients, such as ar-terial blood gas analysis and serum urea nitrogen, werenot available in the consultingoffices of our primary carephysicians.

We tried to find markers for the assessment of theseverity of LRTI among elderly patients in primary care.These 950 home-living LRTI patients with a clinical sus-picion of pneumonia were encountered by primary carephysicians in their everyday practice, and the ongoingstudy did not influence on their treatment decisions. Atleast 3 parameters independently associated with mor-tality within 30 days are easily available: patient history(preceding aggravation of a coexisting illness); physicalfindings (respiratory rate25/min); and laboratory mea-surement (CRP level100 mg/L). Whether these mark-ers are generally applicable to the evaluation of the se-verity and treatment decisions of LRTI among elderlypatients in primary care should be ascertained and vali-dated in a new prospective study.

Accepted for publication March 29, 2001.Corresponding author andreprints: Hannu Syrjala,MD,

PhD, Department of Infection Control, Oulu University Hos-pital, FIN-90220 Oulu, Finland (e-mail: [email protected]).

REFERENCES

1. Macfarlane J. Lower respiratory tract infection and pneumonia in the commu-

nity.Semin Respir Infect. 1999;14:151-162.2. JokinenC, Heiskanen L, JuvonenH, et al.Incidenceof community-acquiredpneu-

monia in the population of four municipalities in eastern Finland. Am J Epide-

miol. 1993;137:977-988.

3. Martson BJ, Plouffe JF, File TM, et al. Incidence of community-acquired pneu-

monia requiring hospitalization: results of a population-based active surveil-

lance study in Ohio.Arch Intern Med. 1997;157:1709-1718.

4. May DS, Kelly JJ, Mendlein JM, Grabe PL. Surveillance of major causes of hos-

pitalizationamong the elderly,1988. MorbMortal WklyRep CDCSurveillSumm.

1991;40:7-21.

5. Dixon RE. Economic cost of respiratory tract infections in the United States.Am

J Med. 1985;78(suppl 68):45-51.

6. Guest JF, Morris A. Community-acquired pneumonia: the annual cost to the na-

tional Health Service in the UK. Eur Respir J. 1997;10:1530-1534.

7. ColeyCM, LiY, Medsger AR,et al.Preferencesfor home vs hospitalcare among

low risk patients with community-acquired pneumonia.Arch Intern Med. 1996;

156:1565-1571.

8. Bartlett JG, Breiman RF, Mandell LA, File TM Jr. Community-acquired pneumo-

nia in adults: guidelines for management. Clin Infect Dis. 1998;28:811-838.

9. The British Thoracic Society and the Public Health Laboratory Service.

Community-acquired pneumonia in adults in British hospitals in 1982-1983: a

survey of aetiology, mortality, prognostic factors and outcome.Q J Med. 1987;

62:195-220.

10. Karalus NC, Cursons RT, Leng RA, et al. Community-acquired pneumonia: ae-

tiology and prognosis index evaluation.Thorax. 1991;46:413-418.

11. Farr BM,Sloman AJ,Fisch MJ. Predictingdeath in patients hospitalized for com-

munity-acquired pneumonia.Ann Intern Med. 1991;115:428-436.

12. American Thoracic Society. Guidelines for the initial management of adults with

community acquired pneumonia: diagnosis, assessment of severity, and initial

antimicrobial therapy.Am Rev Respir Dis. 1993;148:1418-1426.13. Ewig S, Bauer T, Hasper E, Pizzulli L, Kubini R, Luderitz B. Prognostic analysis

and predictive rule for outcome of hospital-treated community-acquired pneu-

monia.Eur Respir J. 1995;8:392-397.

14. Neill AM, Martin IR, Weir R, et al. Community-acquired pneumonia: aetiology

and usefulness of severity criteria on admission. Thorax. 1996;51:1010-1016.

15. PorathA, SchlaefferF, LiebermanD. Appropriateness of hospitalization of patients

with community-acquired pneumonia.Ann Emerg Med. 1996;27:176-183.

16. 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;336:243-250.

17. Conte HA, Chen YT, Mehal W, Scinto JD, Quagliarello VJ. A prognostic rule for

elderlypatients admitted with community-acquired pneumonia.AmJMed. 1999;

106:20-28.

18. Dean NC. Use of prognostic scoring and outcome assessment tools in the ad-

mission decision for community-acquired pneumonia. Clin Chest Med. 1999;

20:521-529.

19. Honkanen PO, Herva E, Kestinen T, et al. Incremental effectiveness of pneumo-

coccal vaccine on simultaneously administered influenza vaccine in preventing

pneumoniaand pneumococcalpneumoniaamongpersonsaged 65 years orolder.

Vaccine. 1999;17:2494-2500.

20. Niederman MS, Fein AM. Pneumonia in the elderly. Clin Geriatr Med. 1986;2:

241-268.

21. Gravelyn TR, Weg JG. Respiratory rate as an indicator of acute respiratory dys-

function.JAMA. 1980;244:1123-1125.

22. Fine MJ, Smith MA, Carson CA, et al. Prognosis and outcomes of patients with

community-acquired pneumonia: a meta-analysis.JAMA. 1996;275:134-141.

23. Hansson L-O, Lindquist L. C-reactive protein: its role in the diagnosis and fol-

low-up of infectious diseases.Curr Opin Infect Dis. 1997;10:196-201.

24. Gabay C, Kushner I. Acute-phase proteins and other systemic responses to in-

flammation.N Engl J Med. 1999;340:448-454.

25. Kuikka A, Syrjanen J, Renkonen O-V, Valtonen VV. Pneumococcal bacteraemia

during a recent decade.J Infect. 1992;24:157-168.

26. Leroy O, Bosquet C, Vandenbussche C, et al. Community-acquired pneumonia

in the intensive care unit: epidemiological and prognosis data in older people.

J Am Geriatr Soc. 1999;47:539-546.

27. Georges H, LeroyO, VandenbusscheC, et al. Epidemiological features and prog-

nosis of severe community-acquired pneumococcal pneumonia. Intensive Care

Med. 1999;25:198-206.

28. Brancati FL, Chow JW, Wagener MM, Vacarello SJ, Yu VL. Is pneumonia really

the old mans friend? two-year prognosis after community-acquired pneumo-

nia.Lancet. 1993;342:30-33.

29. Rello J, RodriguezR, JubertP, AlvarezB, andthe study group.Severecommunity-

acquired pneumonia in the elderly: epidemiology and prognosis.Clin Infect Dis.

1996;23:723-728.

30. Riquelme R, TorresA, El-EbiaryM, et al. Community-acquired pneumoniain the

elderly: a multivariate analysis of risk and prognostic factors. Am J Respir Crit

Care Med. 1996;154:1450-1455.

31. Ewig S, Kleinfield T, Bauer T, Seifert K, Scha fer H, Goke N. Comparative valida-

tion of prognostic rules for community-acquired pneumonia in an elderly popu-

lation.Eur Respir J. 1999;14:370-375.



Documents

2001 - Severity Assessment of Lower Respiratory Tract Infection in Elderly Patients in Primary Care