Allergy 2002: 57: 107–114 Copyright # Munksgaard 2002

Printed in UK. All rights reserved

ALLERGY
ISSN 0105-4538

Original article

The score for allergic rhinitis (SFAR): a simple and valid assessment
method in population studies

Background: No validated assessment of allergic rhinitis (AR) is presently I. Annesi-Maesano1, A. Didier2,

available that can be used in population studies in the absence of medical M. Klossek3, I. Chanal4, D. Moreau1,
diagnosis and of objective measurements of allergy. To compensate for this lack, J. Bousquet5
a quantitative Score For Allergic Rhinitis (SFAR) ranging between 0 and 16 has 1
INSERM U472, Villejuif; 2Hôpital Rangueil, Toulouse;
3
been developed by experts. CHU, Poitiers; 4Hôpital Rotschild, Paris; 5Hôpital
Methods: The SFAR, encompassing eight features of AR, was validated in three Arnaud de Villeneuve, Montpellier, France
different ways: 1) among 269 outpatients taking the specialist’s diagnosis of AR
and skin prick tests (SPT) positivity as a gold standard (diagnosis validation); 2)
using psychometric methods (internal validation); and 3) in a random
population-based sample of 3001 individuals by telephone interview (population
acceptability).
Results: A SFAR value i 7 allowed satisfactory discrimination between the
outpatients with AR from those without (sensitivity=74% [95% confidence
interval CI: 0.69,0.79], specificity=83% [0.79, 0.87], positive predictive
value=84% [0.80, 0.88], negative predictive value=74% [0.69, 0.79] and Key words: acceptability; allergic rhinitis; diagnosis;
Youden’s index=0.57, respectively). Internal consistency of the score was also identification; validation.
high (among others, Cronbach’s alpha coefficient=0.79). On average, it took
I. Annesi-Maesano
only 3 min for the individuals interviewed on the phone to complete the INSERM U472
questionnaire, the questions of which were well understood. Among these Epidemiology & Biostatistics
subjects, the prevalence of AR was 21% [95% CI: 19.5%, 22.5%], which is 16 Ave P.V.-Couturier
comparable to other determinations in France. F94807 Villejuif CEDEX
Conclusions: The newly a priori proposed Score For Allergic Rhinitis (SFAR) is France
easy to use and can be useful to estimate prevalence and to study causation of AR
in population settings. Accepted for publication 10 September 2001

Allergic rhinitis (AR) is a common condition affecting European Community Respiratory Health Study
5–40% of the general population and there is evidence (ECRHS) studies (5) suggests that questions on
that its prevalence is rising (1). An important contribu- hayfever are relevant. However, due to the type of
tion to the investigation of AR is provided by construct design used, the ISAAC questionnaire does
epidemiological studies conducted in large popula- not compute any quantitative score for AR.
tion-based samples. In such samples (where it is difficult Nevertheless, it has been shown that quantitative
to obtain doctor’s evidence of AR or to identify the scores are more informative than dichotomous vari-
environmental antigens that are responsible for AR) ables in characterization of the disease (6).
‘working definitions’ of AR must be implemented in A working party of six specialists was convened in
order to discriminate with reasonable accuracy between France to develop a minimum list of reliable questions
those individuals who have AR and those individuals about AR, which could be used to compute a
who are free from it. A quite complete working quantitative diagnostic criteria score for differentiation
definition of allergic rhinitis among children drawn between AR and other nasal problems (such as
from population-based samples has been provided, infectious rhinitis, occupational rhinitis, drug-induced
using a set of six questions, by the International Study rhinitis, hormonal rhinitis, idiopathic rhinitis) in the
of Asthma and Allergies in Childhood (ISAAC) (2). absence of a doctor’s diagnosis. The score was
The ISAAC questionnaire for rhinitis has been developed to separate AR according to its variety
validated both in children (3) and adults (4) using (allergen, season of the year, etc.), information which is
skin prick tests (SPT), and the fact that the prevalence greatly relevant in the study of AR. In order to assess
distribution of hayfever was similar in the ISAAC and the validity of the proposed Score For Allergic Rhinitis

107
Annesi-Maesano et al.

(SFAR), three methods were used. Firstly, the SFAR Validation of the score
was validated using a doctor’s diagnosis of AR among Three validations of the SFAR were performed: diagnosis validation,
outpatients (diagnostic validation). Secondly, internal internal validation and population acceptability.
validation of items was performed using psychometric In the diagnosis validation, the gold standard diagnosis of rhinitis
methods. Thirdly, the score acceptability was tested in a was made by the specialist’s evaluation and the results of SPT were
used to assess the SFAR cut-off point value discriminating
random population-based sample. individuals with AR from those without. The reliability of the
specialists in diagnosing allergic rhinitis was guaranteed by a
common clinical definition of the disease, according to European
consensus guidelines for the diagnosis and treatment of allergic
Material and methods rhinitis (7).
Study population Internal validation of the SFAR required a quantitative approach
consisting of the study of the items’ validity (content validity, face
Data were obtained from 269 consecutive first-visit outpatients validity, construct validity) and homogeneity, respectively. Content
during 1 month in 1997 referred by the GP because of nasal validity, the extent to which the considered items incorporated the
problems to the Allergy or ENT clinics of four hospitals situated in domain of AR, was guaranteed by the expertise of the panel. This
different zones of France (Brest, Montpellier, Poitiers, and was also confirmed by previously published data (1). Face validity of
Toulouse). Each outpatient filled in a questionnaire containing the items was given by the high degree of acceptability and
items on nasal problems and related features (see Appendix A) and comprehension of the outpatients, in particular those of a pilot
was then examined by a clinical consultant who was blinded to the study conducted separately. Construct validity and homogeneity
answers of the questionnaire. The consultant prescribed SPT to were tested as presented in the section Statistical methods.
major aeroallergens when he/she thought that it was necessary to Finally, the population acceptability, defined as the degree of
ascertain atopic status of the outpatient. Standard aeroallergens were comprehension of the score among individuals drawn from the
used to this end. general population, was tested in a population-based sample by
A random population-based sample of 3001 individuals aged telephone interview. Actually, face validity had been assessed
i 18 years, representing the general adult population of 22 regions amongst only outpatients of the clinics. The sample was randomly
of France, participated by phone in the study of population selected using the phone books of 22 regions of metropolitan France
acceptability of SFAR. The sample consisted of consecutive after having estimated with appropriate techniques the different
individuals at home during the phone call, until the estimated sample sizes. It was decided that in the case of nonresponse, the next
sample size was reached for each region. Appropriate methodology phone number on the list would be tried. This survey provided an
was applied to estimate the size and to make the sample opportunity to estimate the prevalence of allergic rhinitis in France.
representative. Phone numbers for the different regions were
randomly selected from the phone directories that were available
on computer files. Statistical methods
Diagnosis validation. Using the key medical diagnosis of AR as a
gold standard, the following four different methods (8), largely
Development of the score
applied in clinical epidemiology, were considered to assess the
The epidemiological questionnaire on nasal problems (see Appendix SFAR cut-off point value (i.e. ‘the best score value’ of the SFAR)
A) included the following items as established by the panel of allowing optimum discrimination between individuals with AR
specialists in order to construct the quantitative score for AR, the and those without:
SFAR: 1. sensitivity, that is the proportion of individuals with AR who
1. nasal symptoms in the past year, including sneezing, runny were diagnosed by the doctor; and specificity, that is the proportion
nose, and blocked nose when the subject did not have a cold or ‘flu’, of individuals without AR who were not diagnosed (sens and spec,
in the past year; respectively);
2. nasal symptoms accompanied by itchy-watery eyes (rhino- 2. positive and negative predictive values: PPV, the probability
conjunctivitis); of AR, given the results of the tests (diagnosis and SPT); and NPV,
3. months of the year in which nasal symptoms occur. Seasonal the probability of not having AR, given the results of the tests,
(pollen season) vs. perennial rhinitis could then be assessed according respectively;
to the pollen calendar of each region; 3. receiver operator characteristics (ROC) curve, which is used to
4. triggers of nasal symptoms including pollens, house dust describe the accuracy of the SFAR over a range of cut-off values and
mites, house dust and epithelia; thus helps to decide where the best cut-off value of the score would
5. perceived allergic status; be;
6. previous medical diagnosis of allergy; 4. the Youden’s index (i.e. J=specificity+sensitivityx1), which
7. previous positive tests of allergy; allows comparison of SFAR values in terms of specificity and
8. familial history of allergy. sensitivity under the assumption that specificity and sensitivity have
For each item a weighting was attributed by the working party of to be equally important.
specialists on the basis of their experience in clinical practice and
using appropriate statistical methods .(Table 1). Specialists’ choice
was confirmed by medical knowledge on characteristics of rhinitis Internal validation. Construct validity (9) of the SFAR items was
(1). Statistical methods included: multiple correspondence analysis examined using the following complementary methods (6):
(see under Statistical methods) which allowed lining up and thus 1. The multicorrespondence analysis (MCA) (10). This is an
weighting the various items according to the presence or the absence explanatory technique for visually interpreting multivariate data. It
of a medical diagnosis of AR; and odds ratios (OR) between medical involves three or more categorical variables and graphical display of
diagnosis of AR and the various items estimated using the logistic the corresponding contingence tables.
regression model. 2. The graphical method of analysis of the line of items.
For each outpatient, the SFAR based on the number of points According to this method the more the total score is elevated the
made in the questionnaire was then set. The SFAR ranged between 0 more the percentage of positive answers to one item must increase
and 16. (11).

108
 The Score For Allergic Rhinitis (SFAR)

Table 1. Self-completed questionnaire, attributed score and repartition of the items for the Score For Allergic Rhinitis (SFAR)

Outpatients reporting Outpatients with medical diagnosis

Score (points) Cumulative the item (%) of AR reporting the item
Items/discriminators attributed by experts score (n=269) (n=144)

Blocked nose, runny nose, sneezing in past year (nasal symptoms) 1 for each symptom 3 83.61 95.81
Months of the year 1 for perennial 5 43.5 50.0
1 for pollen season 41.6 51.4
Nasal symptoms plus itchy eyes (rhinoconjunctivitis) 2 7 44.6 61.8
Triggers:*
Pollens, house dust mites, dust 2 46.82 69.42
Epithelia (cat, dog) 1 9 15.2 23.6
Perceived allergic status 2 11 51.3 75.0
Previous positive allergic tests 2 13 35.3 52.8
Previous medical diagnosis of allergy 1 14 29.0 38.2
Familial history of allergy 2 16 25.7 38.2
Total points 16

*Divided according to factor analysis.

1
At least one symptom.
2
At least one trigger.

Homogeneity of the SFAR items was established using: population acceptability study, classes of age were equal-
1. The alpha coefficient of Cronbach (11). It has been established ly represented but there were slightly more men than
that to a coefficient included in the interval [0.61, 0.80] correspond a
good agreement among items used to compute it. women. The predominant occupations were managers,
2. The focused principal component analysis (focused FPCA) teachers and being retired.
(6). This method allows description of the relationships among
variables by prioritising the relationships of a particular variable
(medical diagnosis of allergic rhinitis, in our case) to the others (the
Diagnosis validation
items), by computing the correlations of this variable with the others. Sensitivity and specificity as well as predictive values
Lastly, the weights attributed to the SFAR items were obtained indicated that a SFAR cut-off value i 7 optimally dis-
statistically on the basis of the results of both MCA and the logistic
regression model in which the medical diagnosis of AR was included
criminated between individuals with AR and those with-
as a dependent variable and the items as independent variables (10). out .(Figs 1,2 and 3). This value allowed the SFAR to be
Analyses were performed using BMDP software (University of sufficiently sensitive (74%, [95% CI : 0.69, 0.79]) and
California Press, Berkeley, 1992) (10). specific (83%, [0.79, 0.87]) and to have good positive
(84%, [0.80, 0.88]) and negative (74%, [0.69, 0.79]) pre-
dictive values, respectively. Similarly, expressing the
Results relationship between specificity and sensitivity with a
Characteristics of the study samples ROC curve showed that the best SFAR cut-off value
was i 7, as this value crowded individuals toward the
One hundred and forty-one males and 128 females upper left corner of the ROC curve .(Fig. 4). Further-
participated in the hospital validation study. There was more, a diagnostic SFAR of i 7 was derived on the
no difference in age among them. Main occupations basis of the Youden’s index .(Fig. 5). Results were con-
were teachers, clerks and technicians. The distribution firmed when excluding individuals who had not per-
of each item in these outpatients is shown in Table 1. formed SPT (sensitivity=88% [0.83, 0.93], specificity=
Out of the initial 269 individuals, 144 (53%) were 69% [0.62,0.77], PPV=84% [0.78, 0.90], NPV=76%
diagnosed as suffering from allergic rhinitis at the time [0.69, 0.83] and J=0.57, respectively). It must be noted
of the survey during the consultation .(Fig. 1). One that the weightings attributed by the specialists were
hundred and forty-six outpatients (54%) had SPT to confirmed using the logistic regression models as well as
common aeroallergens following the physician’s request the MCA (data not shown). The more the item was
for allergic tests. Seventy-three percent of the subjects weighted in the score, according to the specialists’ point
having performed SPT were found to be positive to at of view, the higher the corresponding OR was. Further-
least one aeroallergen. Eighty-nine individuals (61%) more, the items were ranked in the same order of import-
had both a clinical diagnosis of AR and positive SPT; ance by MCA.
34 (23%) did not present any positivity and were not
diagnosed as suffering from AR; 17(11%) presented
positive SPT but were not diagnosed as suffering from Internal validation
AR (because of positivity to allergens other than those Regarding construct validity of the SFAR, the MCA
usually involved in AR); and 6 (4%) were diagnosed as showed that on the first axis (32.6% of the inertia) all the
suffering from AR but had negative SPT. affirmative answers to the items were close to the
Among the 3001 individuals that participated in the doctor’s diagnosis of AR and, as expected, on the

109
Annesi-Maesano et al.

are young age, living in the north and south-east of

France, working in craft or trade, and being unem-
ployed.

Discussion
In spite of the fact that allergic rhinitis is an extremely
common health problem in industrialized countries, few
standardized instruments have been developed to assess
it. The diagnostic criteria of SFAR, as proposed herein,
are easy to use; SFAR takes under 3 min per individual
to be complete and might be useful to estimate AR
prevalence in population settings, as well as to study
causation of AR.
The landscape offered by instruments allowing the
assessment of allergic rhinitis has been poor until now.
Most instruments deal with quality of life and/or treat-
ment benefits among individuals with rhinitis. A pref-
erence-based measure of rhinitis symptoms, useful for
clinical trials and for cost-effectiveness studies compar-
ing treatments for rhinitis, is the Rhinitis Symptom
Utility Index (12), which consists of 10 questions on the
severity and the frequency of stuffy or blocked nose,
runny nose, sneezing, itching, watery eyes and itching
nose or throat over a 14-day period. A simple symptom
scale assessing rhinoconjunctivitis and asthma, two
Figure 1. Study design and repartition of the outpatients interdependent conditions, was developed and fully vali-
according to allergic rhinitis (AR) and skin prick test (SPT) dated in 102 patients with rhinoconjunctivitis and asth-
positivity (n=269) ma (13). The purpose was to use these two instruments
with existing validated tools such as treatment needs and
quality of life assessment, in order to provide a compre-
second axis (12.4% of inertia) seasonal rhinitis was hensive picture of allergic airway disease for quality
opposed to perennial rhinitis. The analysis of the lines of assurance or research purposes. A 134-item question-
the items always showed unimodal curves except in the naire was developed to distinguish between patients with
cases of seasonal and perennial rhinitis and familial allergic rhinitis (AR) and those with perennial nonaller-
history of allergy. Regarding homogeneity, the internal gic rhinitis or vasomotor rhinitis (VMR) (14). Although
consistency of the SFAR was good (Cronbach’s alpha= useful in pointing out the many differences in the med-
0.79), suggesting that all the items were appropriate for ical histories of AR and VMR patients, this question-
assessing AR. Furthermore, the FPCA focusing on the naire is not practical in epidemiological studies. Two
medical diagnosis of AR showed that previous medical instruments were used in population studies to assess
diagnosis of allergy, perceived allergic status, previous AR. The ISAAC core questionnaire for rhinitis consist-
positive tests of allergy, rhinoconjunctivitis, and being ing of six questions aimed at distinguishing between
susceptible to the various triggers (house dust mites, rhinitic and nonrhinitic individuals in the general
pets, etc.) constituted a group of variables that corre- population, predicting which subjects with rhinitis are
lated positively with each other (data not shown). likely to be atopic and giving some indication of the
severity of rhinitis among affected individuals. ISAAC
questions on nasal symptoms when the subjects did not
have a cold or flu were found to have a positive predict-
General acceptability
ive value of 80% in detecting rhinitis in a community
The questionnaire was well understood by individuals sample of adults (aged 16–65 years) (15). The ISAAC
randomly drawn from the general population, who question on rhinoconjunctivitis had the highest positive
were interviewed over the phone. There was no effect of predictive value (78%) for detecting atopy among
age, sex or socioeconomic status in the answers. Among subjects with rhinitis in the same population. Finally,
the 3001 individuals interviewed, 21% [95% CI: 19.5%, the ISAAC question on the month of the year during
22.5%] had a SFAR i 7. Factors associated with a which symptoms occur had a positive predictive value of
population with AR, as defined on the basis of SFAR, 71% in detecting atopy among subjects with rhinitis. In

110
 The Score For Allergic Rhinitis (SFAR)

Figure 2. Sensitivity and specificity of the Score For Allergic

Rhinitis (SFAR) (n=269)

the a posteriori validation of the ISAAC core questions Figure 3. Positive (+) and negative (-) predictive values of the
on rhinitis, by comparing them with SPT results in a Score For Allergic Rhinitis (SFAR) (n=269)
large population-based sample of Swiss schoolchildren
(3), questions on rhinitis were found to be highly specific between 16 and 82 years of age (16). However, this
(ranging from 77.5% to 97.6%) and therefore useful for questionnaire, for which few elements of validation were
excluding atopy. In addition, they had a high positive provided, has not been used in other studies. Recently, a
predictive value (63% for sneezing accompanied by scoring system was introduced to diagnose allergic
itchy-watery eyes, 67% for symptoms occurring only rhinitis on a routine basis in a small cohort of 47 allergic
during the pollen season and 70% for reported hayfever) rhinitis and 23 normal subjects assembled at the hospital
in detecting atopy among subjects with symptoms. (17). However, the scoring was based on symptoms,
However, they were not helpful for detecting atopy in signs, SPT and IgE levels, and is therefore difficult to use
the general population (low sensitivity). By question- in large populations. Similarly, questions on allergic
naire, both allergic and nonallergic nasal complaints rhinitis were validated using objective measurements of
were assessed among 1469 randomly selected persons SPT and specific IgE (18) but in 150 subjects who had

Figure 4. Receiver operator characteristics (ROC) curve of the Score For Allergic Rhinitis (SFAR) (n=269)

111
Annesi-Maesano et al.

nasal symptoms was taken into account, as the

Cronbach’s coefficient was always higher than 0.70.
Reporting in detail the months in which nasal
symptoms occurred has not been shown to provide
much better information because of recall bias.
Actually, individuals are prone to report that nasal
symptoms occurred in months of the year closer to the
period of the survey both in France (19) and elsewhere
(20). That is why we did not use the SFAR to
discriminate between seasonal and perennial rhinitis.
However, this could be done after further implementa-
tion and development of specific questions. More
Figure 5. Youden’s index according to the different values of the relevant information on the variety of AR can be
Score For Allergic Rhinitis (SFAR) (n=269) obtained by considering the allergens to which the
individual is sensitized. From a practical point of view,
ever reported asthma or wheezing and 140 without a questionnaire in large scale population surveys will be
symptoms in Finland so attributing a special meaning to successful if it is simple and acceptable to the subjects,
the validation. requiring only a few minutes to complete, if it has
Compared to all the other existing instruments, the adequate reproducibility and yields relevant informa-
SFAR developed here has two advantages. From the tion which cannot be obtained more conveniently in
methodological point of view, the SFAR is a quantita- other ways. The SFAR seems to satisfy all such criteria
tive score, thus more informative than the other unidi- but intraindividual reproducibility has not been tested.
mensional instruments usually employed. This meant The SFAR is a very short validated instrument
that better sensitivity, specificity and predictive values including only eight items which can be assessed with
were obtained than with other instruments were obtain- a self-reported questionnaire taking only 3 min to be
ed in ascertaining individuals’ suffering from AR. The filled in. The phone survey showed that the SFAR is
SFAR has a higher positive predictive value and well accepted by individuals of various socioeconomic
specificity than the ISAAC questionnaire on rhinitis. statuses.
More importantly, the SFAR has a much higher In conclusion, for the first time a quantitative score
sensitivity than the ISAAC questionnaire, which has for AR (SFAR) is proposed that can be useful in
been shown to be unhelpful in the detection of atopy in estimating prevalence and to study causation of AR in
the general population of children. Excluding indivi- population settings. Additional uses of the SFAR might
duals not having performed SPT and taking SPT include the determination of AR probands in genetic
positivity into account yielded even better results in studies which aim to identify genetic markers of this
terms of sensitivity (88% vs. 74%) in the case of SFAR. disease.
Furthermore, appropriate statistical methods issued
from psychometric techniques were applied for the
internal validation of SFAR. This is not usually done in
the development of a score for AR (13). Psychometric Acknowledgments
techniques indicated that all the items were very good. This work was partly supported by Rhône–Poulenc Rohrer (France).
The SFAR was homogeneous even when seasonality of

References
1. JOHANSSON SGO, HOURIHANE JO’B, 3. BRAUN-FAHRLANDER C, WUTHRICH B, 4. CHARPIN D, SIBBALD B, WEEKE E,
BOUSQUET J, BRUIJNZEEL-KOOMEN GASSNER M, et al. Validation of a WUTHRICH B. Epidemiologic
DREBORG S, HAATELA T, et al. A revised rhinitis symptom questionnaire (ISAAC identification of allergic rhinitis. Allergy
nomenclature for allergy. Allergy core questions) in a population of Swiss 1996;51:293–298.
2001;56:813–824. school children visiting the school 5. PEARCE N, SUNYER J, CHENG S, et al.
2. ASHER MI, KEIL U, ANDERSON HR, et al. health services. SCARPOL team. Swiss Comparison of asthma prevalence in the
International Study of Asthma and Study on Childhood Allergy and ISAAC and the ECRHS. Eur Resp J
Allergies in Childhood (ISAAC): Respiratory Symptom with respect to 2000;16:420–426.
rationale and methods. Eur Respir J Air Pollution and Climate. 6. FALISSARD B. Focused principal
1995;8:483–491. International Study of Asthma and component analysis: looking at a
Allergies in Childhood. Pediatr Allergy correlation matrix with a particular
Immunol 1997;8:75–82. interest in a given variable. J
Computational Graph Stat 1999;8:12.

112
 The Score For Allergic Rhinitis (SFAR)

7. International Consensus Report on the 12. REVICKI DA, LEIDY NK, BRENNAN- 17. NG MLS, WARLOW RS, CHRISHANTHAN
diagnosis and management of rhinitis. DIEMER F, THOMPSON C, TOGIAS A. N, ELLIS C, WALLS S. Preliminary
International Rhinitis Management Development and preliminary criteria for the definition of allergic
Working Group. Allergy 1994;49:1–34. validation of the multiattribute Rhinitis rhinitis: a systematic evaluation of
8. FLETCHER RH. Clinical Epidemiology. Symptom Utility Index. Qual Life Res clinical parameters in a disease cohort
The Essentials, 2nd edn. Baltimore: 1998;7:693–702. (II). Clin Exp Allergy
Williams & Wilkins, 1988. 13. WASSERFALLEN JB, GOLD K, SCHULMAN 2000;30:1417–1422.
9. LAST JM. A Dictionary of KA, BARANIUK JN. Development and 18. KILPELÄINEN M, TERHO EO, HELENIUS H,
Epidemiology, 3rd edn. New York: validation of a rhinoconjunctivitis and KOSKENVUO M. Validation of a new
Oxford University Press, 1995, p.171. asthma symptom score for use as an questionnaire on asthma, allergic
10. XXX X BMDP Statistical Software outcome measure in clinical trials. J rhinitis, and conjunctivitis in young
Manual, Vol. 2. Los Angeles,:, Allergy Clin Immunol 1997;100:16–22. adults. Allergy 2001;56:377–384.
University of California Press, 1992, 14. LINDBERG S, MALM L. Comparison of 19. ANNESI-MAESANO I, ORYSZCZYN MP. La
733–754, , 1105–, 1143. allergic rhinitis and vasomotor rhinitis rhinite de l’adolescent. Résultats
11. MORET L, MESBAH M, CHWALOW J, patients on the basis of a computer L’enquête ISAAC. Rev Fr Allergologie
LELLOUCH J. Internal validation of a questionnaire. Allergy 1993;48:602–607. 1998;38:283–289.
measurement scale: relation between 15. SIBBALD B, RINK E. Epidemiology of 20. STEWART AW, ASHER MI, CLAYTON TO
principal component analysis, seasonal and perennial rhinitis: clinical et al. The effect of season-response to
Cronbach’s alpha coefficient and intra- presentation and medical history. ISAAC questions about asthma, rhinitis
class correlation coefficient. Rev Thorax 1991;46:895–901. and eczema in children. Int J Epidemiol
Epidemiol Sante Publique 16. JESSEN M, JANZON L Prevalence of non- 1997;26:126–136
1993;41:179–186. allergic nasal complaints in an urban
and a rural population in Sweden.
Allergy 1989;44:582–587.

113
Annesi-Maesano et al.

Appendix

Standardized questionnaire for SFAR assessment

1 In the past 12 months, have you had a problem apart from cold or flu with (please tick appropriate cases(s)) :
Sneezing No u Yes u
Runny nose No u Yes u
Blocked nose No u Yes u

If YES (at least one nose problem):

2 In the past 12 months, has this nose problem been accompanied by itchy-watery eyes?
No u Yes u

3 In which of the past 12 months (or in which season) did this nose problem occur?
Jan u Feb u Mar u Apr u
May u June u July u Aug u
Sept u Oct u Nov u Dec u
(or alternatively)
Winter u Spring u Summer u Autumn u

4 What trigger factors provoke or increase your nose problem?

House dust u
House dust mites u
Pollens u
Animal (cat, dogs . . . ) u
Others (please specify) .............................................................................................................................................................................................................................................................

5 Do you think to be allergic?

No u Yes u

6 Have you already been tested for allergy (SPT, IgE)?

No u Yes u

If YES:
6a What was the result?
Positive u Negative u

7 Has a doctor already diagnosed that you suffer/suffered from asthma, eczema or allergic rhinitis?
No u Yes u

8 Is any member of your family suffering from asthma, eczema or allergic rhinitis?
No u Yes u

If YES: Who and what disease? (please tick appropriate cases(s)) :

Father Asthma u Eczema u Allergic rhinitis u
Mother Asthma u Eczema u Allergic rhinitis u
Siblings Asthma u Eczema u Allergic rhinitis u

114