Content uploaded by Berthold P Gersons
Author content
All content in this area was uploaded by Berthold P Gersons on Dec 05, 2015
Content may be subject to copyright.
Follow-up after Legionnaires Disease •CID 2002:35 (1 July) •11
MAJOR ARTICLE
Health-Related Quality of Life and Posttraumatic
Stress Disorder among Survivors of an Outbreak
of Legionnaires Disease
Kamilla D. Lettinga,
1
Annelies Verbon,
1
Pythia T. Nieuwkerk,
2
Rene E. Jonkers,
3
Berthold P. R. Gersons,
4
Jan M. Prins,
1
and Peter Speelman
1
1
Department of Internal Medicine, Division of Infectious Diseases, Tropical Medicine, and AIDS, and Departments of
2
Medical Psychology,
3
Pulmonology, and
4
Psychiatry, Academic Medical Center, Amsterdam, The Netherlands
A follow-up study of 122 survivors of an outbreak of legionnaires disease (LD) in The Netherlands was
conducted to determine persistence of symptoms, health-related quality of life (HRQL), and presence of
posttraumatic stress disorder (PTSD). Seventeen months after diagnosis of LD, survivors completed a ques-
tionnaire assessing symptoms and HRQL and a questionnaire assessing PTSD. The most prevalent new symp-
toms were fatigue (in 75% of patients), neurologic symptoms (in 66%), and neuromuscular symptoms (in
63%). HRQL was impaired in 7 of the 8 dimensions assessed by the HRQL questionnaire, and 15% of patients
experienced PTSD. Symptoms and impaired HRQL persisted for 11.5 years. As a result of the design of this
study, it could not be inferred whether Legionella pneumophila infection, severe pneumonia in general, or the
outbreak situation was responsible for impaired well-being. However, awareness of this problem by health
care providers may improve the aftercare of patients.
Despite considerable progress in medical and epide-
miologic management of outbreaks of legionnaires dis-
ease (LD), little is known about the effect of such events
on survivors. Two years after the first reported outbreak
of LD, in Philadelphia [1], 58% of the patients still had
not recovered fully [2]. More-detailed knowledge about
recovery after LD is not available.
Long-term outcome after community-acquired pneu-
monia (CAP) can be described in terms of mortality
during the follow-up period and in terms of lung func-
tion [3–5]. An important measure of patient well-being
Received 29 October 2001; revised 7 February 2002; electronically published 7
June 2002.
Financial support: Ministry of Health, The Netherlands (grant CSG/PP, 1074808).
Reprints or correspondence: Dr. Kamilla D. Lettinga, Academic Medical Center,
F4-217, Dept. of Internal Medicine, Div. of Infectious Diseases, Tropical Medi-
cine, and AIDS, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands (K.D.
Lettinga@amc.uva.nl).
Clinical Infectious Diseases 2002;35:11–7
2002 by the Infectious Diseases Society of America. All rights reserved.
1058-4838/2002/3501-0002$15.00
may be health-related quality of life (HRQL) [6, 7]. This
can be assessed by use of either questionnaires about
symptoms or the Medical Outcomes Study Short Form
36-item health survey (SF-36). A study that used a ques-
tionnaire that assessed symptoms found that, for patients
with pneumonia, recovery to a premorbid health status
required 190 days, even for patients with a low severity
of illness at the time of admission to the hospital [6, 7].
In addition, HRQL (as determined by SF-36) returned
to prepneumonia levels within 90 days [7]. Patients who
were admitted to an intensive care unit (ICU) for lung
injury and survived had an impaired quality of life, com-
pared with matched population control subjects, even
after 1–2 years of follow-up [8–10]. A factor that can
contribute to impairment of HRQL is posttraumatic
stress disorder (PTSD), which is characterized by the
development of typical symptoms after the experience
of traumatic events. In ICU survivors, an association
between the development of PTSD and impaired HRQL
has been demonstrated [9].
In 1999, a large outbreak of LD among attendees of
by guest on September 16, 2015http://cid.oxfordjournals.org/Downloaded from
12 •CID 2002:35 (1 July) •Lettinga et al.
Figure 1. Overview of patient population and recruitment criteria for
the study patients with legionnaires disease (LD). “No visitor” means
that the patient did not attend the flower show. “No reaction” means
that the patient did not respond to the request to participate in the study.
a flower show occurred in The Netherlands; 11% of the affected
patients died [11]. This outbreak provided us with the oppor-
tunity to describe the persistence of symptoms and HRQL in
survivors of an outbreak of LD. We also evaluated predictors,
including PTSD, for persistence of symptoms and for impaired
HRQL.
PATIENTS AND METHODS
Study group. In March 1999, an outbreak of LD was detected
in The Netherlands [11]. Local municipal health services and
hospitals were requested to report every suspected case of LD.
Six months after the outbreak took place, 318 patients with
suspected LD had been reported. Written informed consent to
collect clinical data was obtained from 202 patients or their
relatives (figure 1). The study was approved by the medical
ethics committee of the Academic Medical Center in Amster-
dam, The Netherlands.
Patients were considered to have LD if they had visited the
flower show, had experienced symptoms compatiblewith pneu-
monia, had radiologic signs of infiltration, and had laboratory
evidence of Legionella infection. For patients with confirmed
LD, laboratory evidence included the following: (1) isolation
of Legionella pneumophila from a respiratory sputum sample,
(2) detection of L. pneumophila serogroup 1 antigen in a urine
sample (by use of the Binax NOW Legionella urinary antigen
test; Binax), and (3) either seroconversion to positive IgG or
IgM antibody titers or a 4-fold increase in antibody titers to
L. pneumophila in paired acute-phase and convalescent-phase
serum samples, as determined by a commercial ELISA (Serion;
Institut Virion-Serion) or a microagglutination antibody assay
(Regional Laboratory of Public Health, Tilburg). For patients
with probable LD, laboratory evidence included findings of a
single high antibody titer (11:256), a sputum sample positive
for L. pneumophila by PCR, or no laboratory evidence, provided
that no other microorganism was identified. Patients without
radiographic confirmation of pneumonia or without symptoms
of pneumonia were excluded from the study (figure 1). All
patients were treated with antibiotics for 2–3 weeks.
Follow-up. The “post-LD period” was defined as the pe-
riod that began after the date that antibiotic therapy was com-
pleted, which was set at 1 April 1999 for all patients. Patients
with LD were asked to complete 2 sets of questionnaires. The
first set was sent to eligible patients by mail 7 months after the
outbreak. Patients were asked to compare the symptoms they
experienced 2 and 6 months after LD with their premorbid
symptoms. The second set of questionnaires, including the
symptom questionnaire, the SF-36, and the PTSD list, were
sent to the patients by mail or were completed during a hospital
visit related to participation in a study for the evaluation of
residual pulmonary abnormalities; this set of questionnaires
was completed 13–20 months after LD (mean, 17 months).
Definitions. Data on the following variables were collected
from the medical hospital chart or from the general physician:
(1) the patient’s demographic characteristics; (2) presence of
underlying conditions, such as smoking, chronic obstructive
pulmonary disease (COPD), and cardiovascular disease (CVD;
defined as any cardiovascular condition, including hyperten-
sion, that required medication and that was present at the time
of the visit to the flower show); (3) findings of physical ex-
amination, routine laboratory investigations, and chest radi-
ography performed at admission to the hospital and reviewed
by the attending hospital radiologist; and (4) admission to the
ICU and receipt of mechanical ventilation.
LD was classified as “severe” if ⭓2 of the following minor
criteria for severity of pneumonia [12] were present at admis-
sion: (1) respiratory rate of 130 breaths/min, (2) chest radio-
graph showing bilateral involvement or involvement of multiple
lobes, (3) shock (systolic blood pressure of !90 mm Hg or
diastolic blood pressure of !60 mm Hg), and (4) a Pao
2
of !60
mm Hg or an Sao
2
of !92%.
by guest on September 16, 2015http://cid.oxfordjournals.org/Downloaded from
Follow-up after Legionnaires Disease •CID 2002:35 (1 July) •13
Questionnaires. The items on the symptom questionnaire
were selected by a panel of investigators on the basis of previous
studies of LD and CAP [2, 7]. The questionnaire includeditems
on the most prevalent respiratory symptoms (cough and dys-
pnea), fatigue, neurologic symptoms (headache, memory loss,
and concentration problems) and neuromuscular symptoms
(muscle/joint pain, muscle weakness, and tingling in the fingers,
arms, or feet). Symptoms were scored as “present” or “absent”
before LD and at 2, 6, and 17 months after LD.
The SF-36 comprises 36 items that address the following 8
dimensions believed to reflect the respondent’s quality of life:
ability to perform usual and vigorous activities (physical func-
tion), ability to participate in social and occupational activities
(social function, physical role function, and emotional role
function), moods (mental health dimension), amount of energy
and pain (vitality and pain dimensions), and current health
(general health perception) [13]. Each dimension is scored from
0 to 100, with higher scores indicating better quality of life.
Patients’ SF-36 scores were compared with published age- and
sex-matched Dutch reference population norms [14].
Presence of PTSD was measured by a questionnaire with a
self-rating scale. Each item on the questionnaire corresponded
to 1 of the 17 diagnostic criteria for PTSD given in the Di-
agnostic and Statistical Manual of Mental Disorders, Fourth Edi-
tion (DSM-IV). The questionnaire has been validated with use
of a sample of 136 survivors of a plane crash: Cronbach’s a
was .96, indicating excellent internal consistency, and sensitivity
and specificity were 86% and 80%, respectively, compared with
a structured interview (the gold standard) [15]. Items on the
questionnaire are clustered into the following categories: reex-
perience symptoms (5 items), avoidance symptoms (7 items),
and hyperarousal symptoms (5 items). Severity is rated on a
3-point scale, in which a score of 0 indicates “not at all”; a
score of 1 indicates “slightly,” “once,” or “!4 times”; and a
score of 2 indicates “very much,” “almost constantly,” or “⭓4
times.” A symptom is rated as “present” if the score for an
item is ⭓1 on the severity scale or, for some items, ⭓2. PTSD
is diagnosed if at least 1 reexperience symptom, 3 avoidance
symptoms, and 2 hyperarousal symptoms are present. The pa-
tients in our study were not aware that the questionnaire was
used to establish the possible diagnosis of PTSD.
Statistical analysis. Continuous variables were compared
by means of Student’s ttest for groups; categorical variables
were assessed by the x
2
test or Fisher’s exact test. Symptom
resolution 2–17 months after LD was determined by means of
McNemar’s x
2
test. A 2-tailed Pvalue of !.05 was considered
statistically significant.
The 8 dimensions of the SF-36 score were converted to
standard scores on the basis of the scores of an age- and sex-
matched representative reference sample of the Dutch pop-
ulation. Standard scores were calculated by dividing the dif-
ference between the patients’ SF-36 score and the mean score
of the matched reference population by the SDs of the ref-
erence population. A standard score thus indicates how many
SDs the observed SF-36 score falls below or above the score
of the reference population. Consequently, scores of the ref-
erence population are set at 0. Because it is similar to the
effect size calculation, a mean standard score of 0.20 is con-
sidered to indicate a small deviation from the reference pop-
ulation [16], and mean standard scores of 0.50 and 0.80 are
considered to indicate moderate and large deviations from
the reference population, respectively.
Mean standard scores of patients with LD were compared
with those of the Dutch reference population by means of a
1-sample ttest (i.e., by testing whether mean standard scores
from patients with LD were significantly different from zero).
HRQL scores for patients with and for patients without per-
sisting symptoms were compared by means of 1-way analysis
of variance (ANOVA).
RESULTS
Patient characteristics. For the follow-up study, 142 eligible
patients were asked to participate, and 122 patients agreed to
do so (figure 1). Of these 122 patients, 86 participated in the
study that evaluated residual pulmonary abnormalities and also
completed the questionnaires during the same hospital visit.
There were no significant differences betweenthe characteristics
of patients who returned the second set of questionnaires by
mail and the characteristics of those who completed these ques-
tionnaires during the hospital visit; therefore, data for both
groups were combined in the analysis.
The median age of the patients was 66 years (range, 25–87
years); 60% were men, and most patients were hospitalized
(table 1). Thirty-three patients (27%) fulfilled the criteria for
presence of severe LD, 23 patients (19%) survived their stay in
the ICU, and 14 patients (11%) needed to receive mechanical
ventilation. Premorbid conditions, such COPD, diabetes mel-
litus, cancer, or immunosuppression due to disease or medi-
cation, were infrequently reported (in !10% of patients). CVD
was reported in 33% of the patients, and 47% of the patients
were active smokers (⭓1 cigarette per day) at the time of hos-
pital admission.
Questionnaires. The most frequently reported new symp-
toms 2 months after LD were fatigue (reported by 81% of
patients); neurologic symptoms, such as concentration prob-
lems and memory loss (75% of patients); and neuromuscular
symptoms, such as muscle/joint pain or muscle weakness (79%
of patients) (table 2). Coughing (48%) and dyspnea on exertion
(38%) were the most commonly reported new respiratory
symptoms. The prevalence of each symptom decreased at each
subsequent follow-up visit, but a substantial number of symp-
by guest on September 16, 2015http://cid.oxfordjournals.org/Downloaded from
14 •CID 2002:35 (1 July) •Lettinga et al.
Table 1. Characteristics of 122 patients who received a diag-
nosis of legionnaires disease after receiving antibiotic therapy.
Characteristic Value
Confirmed cases 102 (84)
Probable cases 20 (16)
Age, median years (range) 66 (25–87)
Sex
Male 73 (60)
Female 49 (40)
Underlying conditions
a
Smoking 51 (47)
Chronic obstructive pulmonary disease 12 (10)
Cardiovascular disease 39 (33)
Treatment setting
Hospital (inpatient) 105 (86)
Outpatient clinic 7 (6)
General physician 10 (8)
Severe legionnaires disease 33 (27)
Length of hospital stay, median days (range) 12 (3–132)
Admitted to intensive care unit 23 (19)
Received mechanical ventilation 14 (11)
NOTE. Data are no. (%) of patients, unless other wise specified.
a
Data on smoking were available for 109 patients; data on chronic obstruc-
tive pulmonary disease were available for 117 patients; and data on cardio-
vascular disease were available for 118 patients.
Table 2. Symptoms self-reported by patients with legionnaires disease after receipt
of antibiotic therapy, by months after diagnosis.
Questionnaire item
Months after diagnosis Pfor reduction
of symptoms
a
2617
Fatigue 92/113 (81) 84/110 (76) 90/120 (75) .26
Neurologic symptoms
b
85/113 (75) 76/113 (67) 80/121 (66) .0001
Neuromuscular symptoms
c
89/113 (79) 83/113 (73) 76/121 (63) .004
Coughing 51/107 (48) 43/106 (41) 46/118 (39) .15
Shortness of breath
At rest 25/108 (23) 23/108 (21) 19/115 (17) .26
On exertion
d
40/106 (38) 38/106 (36) 31/110 (28) .06
NOTE. Data are no. of patients with symptom/no. of patients who answered the question (%). The
questionnaire for assessment of self-reported symptoms was completed by 113 patients at 2 and 6
months after diagnosis and by 121 patients at 17 months after diagnosis.
a
The decline in the proportion of patients with symptoms from 2 to 17 months after the diagnosis
of legionnaires disease was calculated with McNemar’s x
2
test.
b
Included headache, dizziness, and loss of concentration or memory.
c
Included paresthesiae in hands or feet, muscle pain, or muscle weakness.
d
Defined as shortness of breath during walking.
toms were still reported 17 months after LD. Only the decline
in neurologic and neuromuscular symptoms between 2 and 17
months after LD reached statistical significance (table 2).
Seventeen months after LD, patients had lower scores for 7
of the 8 dimensions of the SF-36 than did a Dutch age- and
sex-matched reference population (figure 2). For these 7 di-
mensions, the deviation of the HRQL score from that of the
reference population was statistically significant ( ). The
P!.01
most severely reduced dimensions were physical role function
(standard score, ⫺0.68), general health (standard score, ⫺0.64),
and vitality (standard score, ⫺0.55). PTSD was present in 18
patients (15%). Only 1 of these patients had actually been
treated for PTSD by a psychiatrist at the time of examination.
In summary, 17 months after the outbreak of LD, the majority
of patients experienced persistent symptoms and impaired
HRQL, and 15% experienced PTSD.
Predisposing conditions for impaired well-being and
PTSD. Because premorbid conditions and traumatizing
events during hospitalization have been associated with im-
paired HRQL and with PTSD [9, 17, 18], we determined
whether persistence of self-reported symptoms, impaired
HRQL, and presence of PTSD was associated with premorbid
conditions (i.e., age 166 years, sex, COPD, CVD, and smoking)
or determinants of severity of disease (i.e., severe LD, ICU
admission, use of mechanical ventilation, and relatively long
hospital stay). Men reported significantly less dyspnea on ex-
ertion than did women (RR, 0.5; 95% CI, 0.3–0.9), and patients
who were active smokers reported more muscular signs than
did nonsmokers (RR, 1.6; 95% CI, 1.2–2.1). Other premorbid
conditions or disease determinants were not associated with
the presence of any of the symptoms 17 months after LD.
Likewise, no significant association was found between severity
of pneumonia or ICU admission and HRQL dimensions. How-
ever, patients with a history of CVD (median age, 73 years;
range, 53–87 years) had significantly more impairment in phys-
ical function, physical role function, general health, and vitality
( ; Student’s ttest) than did patients without a historyP!.05
of CVD (median age, 64 years; range, 25–78 years).
by guest on September 16, 2015http://cid.oxfordjournals.org/Downloaded from
Follow-up after Legionnaires Disease •CID 2002:35 (1 July) •15
Figure 2. Health-related quality of life (HRQL) scores for patients with
legionnaires disease (LD). The mean follow-up period was 17 months
(range, 13–20 months). The questionnaire for assessment of HRQL was
completed by 121 patients. * , patients with LD versus a DutchP!.01
reference population. Standard scores of !0 indicate an HRQL worse
than that of the reference population, and scores 10 indicate better HRQL.
Dotted lines at 0.2 and ⫺0.2, at 0.5 and ⫺0.5, and at ⫺0.8 indicate
a small, moderate, or large deviation from the reference population,
respectively.
Table 3. Risk factors for the development of a posttraumatic
stress disorder (PTSD) in patients with legionnaires disease pre-
viously diagnosed.
Risk factor RR (95% CI)
Age 166 years (median) 0.5 (0.2–1.2)
Age 156 years
a
0.4 (0.18–0.96)
b
Male sex 0.53 (0.23–1.25)
Underlying conditions
Chronic obstructive pulmonary disease 1.8 (0.6–5.5)
Cardiovascular disease 0.8 (0.3–2.2)
Severe legionnaires disease 1.1 (0.4–2.7)
Admission to intensive care unit 2.3 (0.9–5.3)
Mechanical ventilation 2.4 (0.9–6.1)
Length of hospital stay 112 days 1.2 (0.5–2.8)
Therapy delayed 124 h
c
1.5 (0.6–3.7)
NOTE. The questionnaire for assessment of PTSD was completed by 120
patients. Eighteen patients (15%) fulfilled the criteria for PTSD.
a
That is, being above the 25th percentile for age.
b
;x
2
test.P!.05
c
Adequate antibiotic therapy initiated 124 h after admission to the hospital.
Younger patients (those aged !56 years; i.e., those below the
25th percentile for age) were more likely to develop PTSD than
were older patients ( , table 3). Sex, presence of un-Pp.039
derlying conditions, and fulfillment of the criteria for severe
LD were not associated with the development of PTSD.
Relationships among self-reported symptoms, HRQL, and
PTSD. HRQL was significantly lower for patients who re-
ported fatigue, coughing or shortness of breath on exertion,
neurologic symptoms, or neuromuscular symptoms, com-
pared with the patients who did not report these symptoms
(in all cases , ANOVA). In particular, large deviationsP!.05
from population norms for physical function, physical role
function, and general health were found for patients who
reported symptoms. As expected, HRQL was significantly
lower among patients with PTSD than among patients with-
out PTSD for all 8 dimensions assessed by the SF-36 (in all
cases, , by ttest).P!.05
DISCUSSION
We described the symptoms and assessments of well-being for
a large and well-defined group of 122 patients who survived
LD. The most prevalent symptoms persisting 17 months after
LD were fatigue (in 75% of patients), neurologic symptoms (in
66%), and neuromuscular symptoms (in 63%). Respiratory
symptoms, such as coughing (in 39% of patients) and shortness
of breath (in 28%), were also commonly reported. In contrast
to symptoms after CAP, there was little improvement of self-
reported symptoms between 2 and 17 months after LD. HRQL
was significantly impaired for 7 of 8 dimensions assessed by
the SF-36, and PTSD could be demonstrated in 15% of the
patients. Thus, LD can affect health status in a large proportion
of patients for at least 1.5 years after the completion of anti-
biotic treatment for the disease.
Persistence of symptoms, such as fatigue and dyspnea, was
also described in 18 of 31 people 2 years after the LD outbreak
in Philadelphia [2]. Of the age-matched legionnaires who at-
tended the convention but did not meet the clinical epide-
miologic diagnostic criteria [1], only 6% noted fatigue, and 3%
reported dyspnea on exertion. In addition, of patients who had
low-risk CAP [19], 51% reported fatigue and 28% reported
dyspnea up to 3 months after the completion of antibiotic
treatment [6, 7]. The SF-36 indicated that, in these patients,
HRQL returned to prepneumonia levels 3 months after com-
pletion of antibiotic treatment [6]. However, the period of per-
sistence of symptoms may be longer for patients with high-risk
CAP, which suggests that persistence of symptoms and impaired
quality of life in the patients who had had LD might not be
unique to L. pneumophila pneumonia, but might be associated
with severe pneumonia in general.
We could not demonstrate a significant association between
persistence of symptoms or impaired HRQL and either pre-
morbid conditions, severity of pneumonia, delay in antibiotic
treatment, or length of hospital stay. In contrast, Marrie et al.
[17] reported an association between persistence of symptoms
and age, COPD, and treatment with levofloxacin. This differ-
ence might be explained by the duration of follow-up, which
was 6 weeks in the study of Marrie et al. [17] and 17 months
in the present study.
by guest on September 16, 2015http://cid.oxfordjournals.org/Downloaded from
16 •CID 2002:35 (1 July) •Lettinga et al.
Impaired HRQL and persistence of symptoms have also been
demonstrated 1–2 years after ICU admission, especially in pa-
tients with acute lung injury or acute respiratory distress syn-
drome [8, 10, 20, 21]. We could not demonstrate a significant
increase in self-reported symptoms or impaired HRQL in pa-
tients who had had LD and who had been admitted to the ICU
or had received mechanical ventilation, compared with patients
who had not. No association was found between self-reported
symptoms or impaired HRQL (except for the dimension of
physical role function) and the presence of pulmonary abnor-
malities 17 months after LD (R.E.J., unpublished data). This
suggests that the high percentage of patients who required ad-
mission to the ICU and who had pulmonary damage secondary
to L. pneumophila pneumonia does not explain the incidence
of persistence of symptoms in our study population.
Another factor that can contribute to persistence of symp-
toms and impairment of well-being is PTSD. Patients who are
admitted to the ICU with lung injury and survive often report
nonpulmonary symptoms, such as difficulty concentrating and
memory loss [8]. Patients admitted to the ICU are exposed to
stress by systemic infection [22], and staying in the ICU en-
vironment can be a stressful event. According to the definition
of PTSD in DSM-IV, such a person has been exposed to a
traumatic event and meets both of the following 2 criteria: (1)
the person experienced, witnessed, or was confronted with an
event that involved actual or threatened death or serious injury,
or a threat to the physical integrity of self or others, and (2)
the person’s response to the event involved intense fear, help-
lessness, or horror. Life-threatening disease is an example of
such a stressful event [23]. Major impairments in the mental
health dimensions of HRQL have been associated with the
development of PTSD in such patients [9, 18]. In the patients
we studied, we also found an association between admission
to the ICU and development of PTSD, although it was not
statistically significant, probably because of the small number
of patients. HRQL scores were, indeed, lower for patients who
met the criteria for having PTSD.
Unexplained self-reported symptoms were found at similar
frequencies among a group of people who survived a plane
crash into a neighborhood in Amsterdam [24] and among
soldiers who served in the Gulf War [25]. The percentage of
symptomatic soldiers in the Gulf War who had PTSD was
5%–15% [26]. The percentage of patients with PTSD in our
study (15%) is in agreement with estimations that, in the af-
termath of a disaster, 20%–30% of victims develop PTSD [27].
Also, patients who do not meet the criteria that define PTSD
(according to DSM-IV) may experience symptoms such as reex-
perience, avoidance, or hyperarousal. Therefore, we cannot ex-
clude the possibility that having experienced a life-threatening
illness and being part of an outbreak that was widely and fre-
quently covered in the media plays a role in the development
of the observed sequelae. Also, the intense and exaggerated fear
of situations in daily life that have been linked to LD, such as
contact with water in showers and sprinkler installations, which
can lead to renewed infection with L. pneumophila, may play
an additional role in the development of chronic stress and the
subsequent persistence of symptoms.
This study does not answer the question of whether L. pneu-
mophila itself, severe pneumonia in general, or the outbreak of
LD itself is responsible for the impairment of well-being. More-
conclusive answers regarding the specific cause of impairment
would require the use of both matched control subjects with
severe pneumonia and matched control subjects who con-
tracted LD in the absence of an outbreak. Because of the rel-
atively low incidence of LD and the rarity of well-documented
LD outbreaks, a study designed to answer these questions would
not be feasible. However, our results indicate that health care
providers should be aware of the possibility of persisting symp-
toms, impaired HRQL, and high frequency of (partial) PTSD,
especially in an outbreak situation, because patients will seek
help for their complaints. Awareness of this problem will make
it possible to improve health care for such patients.
Acknowledgments
We thank all the hospital clinicians and general practitioners
who requested their patients’ permission to be included in the
initial study, which allowed us to collect clinical data. We thank
G. J. Weverling and P. M. M. Bossuyt (Department of Clinical
Epidemiology and Biostatistics, Academic Medical Center, Am-
sterdam, The Netherlands), for statistical advice, and J. van
Steenbergen (National Outbreak Structure for Infectious Dis-
eases), for initial data collection by the registrar. We also thank
all the patients who were willing to complete thequestionnaires.
References
1. Fraser DW, Tsai TR, Orenstein W, et al. Legionnaires’ disease: descrip-
tion of an epidemic of pneumonia. N Engl J Med 1977; 297:1189–97.
2. Lattimer GL, Rhodes LV, Salventi JS, et al. The Philadelphia epidemic
of legionnaires’ disease: clinical, pulmonary, and serologic findings two
years later. Ann Intern Med 1979; 90:522–6.
3. Brancati FL, Chow JW, Wagener MM, Vacarello SJ, Yu VL. Is pneu-
monia really the old man’s friend? Two-year prognosis after com-
munity-acquired pneumonia. Lancet 1993; 342:30–3.
4. Gea J, Rodriguez-Roisin R, Torres A, Roca J, Agusti-Vidal A. Lung
function changes following legionnaires’ disease. Eur Respir J 1988;1:
109–14.
5. Fine MJ, Smith MA, Carson CA, et al. Prognosis and outcomes of
patients with community-acquired pneumonia: a meta-analysis.JAMA
1996; 275:134–41.
6. Metlay JP, Fine MJ, Schulz R, et al. Measuring symptomatic and func-
tional recovery in patients with community-acquired pneumonia. J
Gen Intern Med 1997; 12:423–30.
7. Metlay JP, Atlas SJ, Borowsky LH, Singer DE. Time course of symptom
resolution in patients with community-acquired pneumonia. Respir
Med 1998; 92:1137–42.
by guest on September 16, 2015http://cid.oxfordjournals.org/Downloaded from
Follow-up after Legionnaires Disease •CID 2002:35 (1 July) •17
8. Weinert CR, Gross CR, Kangas JR, Bury CL, Marinelli WA. Health-
related quality of life after acute lung injury. Am J Respir Crit Care
Med 1997; 156:1120–8.
9. Schelling G, Stoll C, Haller M, et al. Health-related quality of life and
posttraumatic stress disorder in survivors of the acute respiratory dis-
tress syndrome. Crit Care Med 1998; 26:651–9.
10. Davidson TA, Caldwell ES, Curtis JR, Hudson LD, Steinberg KP. Re-
duced quality of life in survivors of acute respiratory distress syndrome
compared with critically ill control patients. JAMA 1999; 281:354–60.
11. Den Boer JW, Yzerman EP, Schellekens J, et al. A large outbreak of
legionnaires’ disease at a flower show, The Netherlands, 1999. Emerg
Infect Dis 2002; 8:37–43.
12. Niederman MS, Mandell LA, Anzueto A, et al. Guidelines for the
management of adults with community-acquired pneumonia: diag-
nosis, assessment of severity, antimicrobial therapy, and prevention.
Am J Respir Crit Care Med 2001; 163:1730-54.
13. Ware JE, Snow KK, Kosinski M, Gandek B. SF-36 health survey: manual
and interpretation guide. Boston, MA: Health Institute, New England
Medical Center, 1993.
14. Aaronson NK, Muller M, Cohen PD, et al. Translation, validation, and
norming of the Dutch language version of the SF-36 health survey in
community and chronic disease populations. J Clin Epidemiol 1998;
51:1055–68.
15. Carlier IV, Lamberts RD, Van Uchelen AJ, Gersons BP. Clinical utility
of a brief diagnostic test for posttraumatic stress disorder. Psychosom
Med 1998; 60:42–7.
16. Cohen J. Statistical power analysis for the behavioural sciences. Mah-
wah, NJ: Erlbaum, 1988.
17. Marrie TJ, Lau CY, Wheeler SL, Wong CJ, Feagen BG. Predictors of
symptom resolution in patients with community-acquiredpneumonia.
Clin Infect Dis 2000; 31:1362–7.
18. Stoll C, Schelling G, Goetz AE, et al. Health-related quality of life and
post-traumatic stress disorder in patients after cardiac surgery and
intensive care treatment. J Thorac Cardiovasc Surg 2000; 120:505–12.
19. 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.
20. Eddleston JM, White P, Guthrie E. Survival, morbidity, and quality of
life after discharge from intensive care. Crit Care Med 2000; 28:2293–9.
21. Brooks R, Kerridge R, Hillman K, Bauman A, Daffurn K. Quality of
life outcomes after intensive care: comparison with a community
group. Intensive Care Med 1997; 23:581–6.
22. Schelling G, Stoll C, Kapfhammer HP, et al. The effect of stress doses
of hydrocortisone during septic shock on posttraumatic stress disorder
and health-related quality of life in survivors. Crit Care Med 1999; 27:
2678–83.
23. Wilson JP. The historical evolution of PTSD diagnostic criteria: from
Freud to DSM-IV. J Trauma Stress 1994; 7:681–98.
24. IJzermans CJ, van der Zee JS, Oosterhek M, Spreeuwenberg P,Kerssens
J, Donker G. Gezondheidsklachten en de vliegramp Bijlmermeer: een
inventariserend onderzoek naar de aard en omvang, volgens mensen
en/of huisartsen aan de ramp gerelateerde klachten [internal report].
Amsterdam: Academic Medical Center, University of Amsterdam,
1999.
25. Unwin C, Blatchley N, Coker W, et al. Health of UK servicemen who
served in Persian Gulf War. Lancet 1999; 353:169–78.
26. Hyams KC, Wignall FS, Roswell R. War syndromes and their evalu-
ation: from the US Civil War to the Persian Gulf War. Ann Intern
Med 1996; 125:398–405.
27. Hidalgo RB, Davidson JR. Posttraumatic stress disorder: epidemiology
and health-related considerations. J Clin Psychiatry 2000; 61(Suppl 7):
5–13.
by guest on September 16, 2015http://cid.oxfordjournals.org/Downloaded from
