ARTICLE
Genetic studies are based on the analyses of genetically
informative families with a well-defined disease phenotype. With the progress
of the Human Genome Project, these studies are becoming more frequent.
The main way used to phenotype diseases is a consultation by a referent
physician. It is efficient if the diseases are rare or complex, or if
the study concerns small groups.
A study was initiated in France by Généthon in March 1996
to search for susceptibility loci and to identify genes of psoriasis,
a frequent, usually well-defined disease, and considered as a complex
genetic disease [1]. This study comprised three steps: a national media
campaign to identify a large number of psoriasis families; a phenotyping
step and a genotyping study. A media campaign using posters (Fig.
1) in the Paris Métro and information in news magazines, radio
and television, encouraged patients with psoriasis to call a toll-free
telephone number if they wished to participate in a genetic study. This
media campaign produced many phone calls: 50,000 patients or relatives
called Généthon, and led to the identification of 3,800
psoriasis families, 502 of whom had at least five, and 108 at least eight
putative affected members most with an apparently autosomal dominant mode
of inheritance.
We encountered many difficulties in confirming or excluding a diagnosis
of psoriasis: families were numerous and geographically dispersed, and
financial and human resources for phenotyping the patients were limited.
Other difficulties are that psoriasis is a frequent disorder with a usually
benign evolution, the clinical symptoms may fluctuate during the patient's
life and a single consultation may underestimate the right number of patients.
After the evaluation of the cases by a telephone call to all members of
the more genetically informative families, we had three ways to confirm
diagnosis of subjects with putative psoriasis: i) visit of all the members
of the families by a dermatologist of Généthon; ii) phenotype
evaluation by phoning the patient's referent physician; iii) evaluation
by mailing a letter to the patient's referent physician. Finally, the
third method was used.
In the present study, we attempted to assess the efficiency of letters
mailed to patients' referent physicians to confirm the clinical diagnosis
of patients participating in a genetic study. Secondly, we evaluated the
letters and the physicians' characteristics which might modify the response
rate in order to improve it.
Subjects and methods
Families and patients
To select genetically informative families (lod score >= 3.3), we
tested the 108 families with at least eight psoriasis patients with the
SLINK simulation program [2]. All the members of the most genetically
informative families were contacted by telephone by a dermatologist to
determine semiological characteristics of the disease. For confirmation
of the diagnosis, letters were sent to the patients' referent physicians,
except in the following cases: patients who had never consulted a physician
for their skin disorder; patients who refused to authorise the sending
of a letter to their referent physician; affected family members who refused
to participate in the study; and families in which a member was a physician
who could confirm the diagnosis.
Letters
Letters were mailed from January 1997 through June 2000 to a referent
physician designated by the patient. The standard letter (type 1) was
modified in June 1997 to simplify the reply form (type 2). Letters were
on letterhead paper from Généthon. In both letters, the
physician was informed of the patient and family's consent to participate
in a study on psoriasis to be performed by Généthon, and
given a toll-free number to contact Généthon dermatologists
in case he/she wished to have more information about the project. The
main difference between the letters was that in the type 1 letter, physicians
were required to write a new letter to respond to questions about the
diagnosis whereas in the second version (type 2), questions were incorporated
into the letter in questionnaire form, and the physician was only required
to return the form (a translated letter is available from the author:
homme@ genethon.fr).
Physicians' characteristics
Data on medical speciality, sex of the physician, length of practice,
influence of summer months (in France, July and August are the two main
months for holidays), type of practice (private or general practice, single-handed
or with other partners, hospital practice) and size and population of
the area in which the practice was located, were evaluated to characterise
the "good" responders. Length of experience of practice was determined
from the Rosenwald l'annuaire des médecins [3], and calculated
from the date of the thesis of the practitioners. Physicians' locations
were established from the same directory [3] and the telephone web site
http://pagesjaunes.fr. The size of the practice town was determined from
the Dictionnaire des communes [4]. Towns were considered as "small"
if the population was less than 10,000 people.
Ethical considerations and statistical analysis
This study was approved by the Ethical Committee of the University Hospital
of Kremlin-Bicêtre, France, in November 1995.
The frequencies of response rates were compared by using *2
test. Values were considered statistically significant at p < 0.05.
Results
Population studied
From the 108 families with at least eight putative cases of psoriasis,
49 families were analysed. They represented 4,203 individuals (mean size
of families: 85.7 persons, range 28-187). After telephone calls by Généthon
dermatologists, 533 questionnaires concerning 762 patients (mean: 15.5/family,
range 1-66) were mailed to 456 physicians: 302 of the letters (56.7%)
were sent to 268 GPs (58.8%), 212 letters (39.8%) to 172 dermatologists
(37.7%), and 12, five and two letters to nine rheumatologists, five paediatricians,
and two medical internists respectively.
A single letter was sent to 391 physicians. Fifty-nine physicians received
two letters, three received three letters, and one received five letters.
Response rate
The global response rate for the 533 letters was 56.5% (Table
I). There was no difference dependent on the year the letter was sent
(data not shown). Among the 66 physicians to whom two or more letters
were sent, 34 (51.5%) always replied, 17 (25.8%) never replied, and 15
(22.7%) replied at least once.
Characteristics of the letter (Table
I)
The type of the letter sent appeared to be the main factor influencing
the response rate. Neither the number of patients included in the same
letter nor the summer months modified it statistically.
Physicians' characteristics (Table
II)
The dermatologists' response rate was more than 70%, compared with 46%
for the GPs. Neither sex, nor length of the physician's experience modified
the response rate. Physicians who practise in hospitals replied more frequently.
The response rate for physicians who are single-handed practitioners is
not statistically different from the response rate of those with two or
more associates. Finally, the town size in which the practice was located
influenced the response rate only for dermatologists (Table
III).
Discussion
Researchers are seeking the help of physicians (GPs and specialists)
by sending more and more surveys, most often to assess physicians' views
and attitudes. The response rates fluctuate as much as 11 to 90% from
published data [5-12]. Many reasons are evoked to explain such a low response
rate: physicians are swamped by the volume of questionnaires; they resent
interference in their activities by outside researchers; some may be uninterested
or opposed to the research method used; some physicians reproach the researchers
for not providing adequate information about the study and the absence
of feedback at the outcome of the studies in which they participated.
The questionnaires may be too lengthy and time-consuming to elicit a reply;
and finally, some physicians expect a financial incentive [5-7]. Studies
have tried to characterise physicians who do not respond to surveys: they
are usually older, more experienced, less well qualified, and have single-handed
practices [6].
However, these surveys generally deal with physicians' views and attitudes.
Our study is quite different, as it concerns the response rate to a letter
requesting confirmation of diagnosis for specific patients who agreed
to participate in a genetic study on psoriasis.
The response rate of the patients' physicians to confirm a clinical
diagnosis of psoriasis in our study was 56.5%, which was a response rate
often reported by other authors. The particularity of our study is the
size of the survey, which makes the further implications more important.
Ethical and financial aspects must be considered, but the type of the
letters sent and the characteristics of the physicians may contribute
to the response rate.
Ethical considerations
Confidentiality is an important point for the patients and their physicians.
It is preserved by professional secrecy. However, physicians are permitted
to provide medical data about their patients for studies which have obtained
patient and ethical (Ethics Committee) approval [13]. So, theoretically,
ethical considerations should not be an obstacle to physician's reply.
However, some practitioners prefer to avoid the use of letters or phone
calls to give medical information about their patients. To bypass this
difficulty, one of the physicians in our study gave the letter to the
patient who sent it to us himself. For logistical reasons, it is difficult
to generalise this practice.
Two important pieces of information were not mentioned in the letter
sent to the physician: 1) that our project had the approval of an Ethics
Committee, and 2) a written copy of the patient's agreement to send the
letter was not included. Although in 1997 ethical and informed consent
issues did not receive the attention they enjoy today, we consider retrospectively
that provision of this information would probably have increased the response
rate.
Financial considerations
Financial incentive is often mentioned as a means of improving the response
rate to postal surveys, to "cover administrative costs" or to "motivate"
physicians [5, 11]. However some studies with monetary reward before and/or
after the reply, reported a response rate (61%) comparable to ours [11].
In our study, we did not offer any financial incentive. Feedback about
the outcome of scientific research in which physicians and their patients
agreed to participate may be more important. A minor financial and convenience
incentive that might increase the response rate should be the inclusion
of a stamped and pre-addressed return envelope. However, studies using
stamped letters have not reported higher response rates (32-61%) [5, 6,
10].
Quality of questionnaires
The quality of letters seems important and our study confirms this point.
Simplicity, clarity, and ease of response may increase the response rate.
Surprisingly [6], the response rate did not decrease if information in
a single letter concerned many patients. So, we could conclude that it
was not the quantity of information and time required but rather the quality
of the letter that may improve the response rate.
Physicians' characteristics
The response rate was higher for dermatologists than for GPs. A higher
specialist response rate has previously been reported for paediatricians
[14]. Several explanations could explain this difference: fewer postal
questionnaires sent to dermatologists; higher interest and better knowledge
of the disease studied; or more frequent availability of office help.
However, we were surprised to observe that dermatologists represented
only 37.7% of patients' referent physicians. Nevertheless, this low rate
is in accordance with previous data that showed that only 3% of psoriasis
patients consult dermatologists [15].
Hospital practitioners reply more often than office practitioners. Scientific
interest, implication in clinical research, availability of assistance
and better knowledge of ethical considerations may explain this higher
response rate. Moreover, hospital practitioners probably see more severe
psoriasis cases, which could explain a higher interest in research on
this disease.
We have no explanation for the higher response rate of dermatologists
who practice in large towns. Better access to medical information and
therefore higher motivation and interest in research may be an explanation.
Older practitioners have been considered as poor respondents [6] but this
was not confirmed in our study. Sex of the physician did not modify the
response rate [10].
Finally, we were very pleased to state that physicians' motivation did
not decrease during the summer months, since France is reputed to be a
country in which everything grinds to a halt in August.
CONCLUSION
A well-defined phenotype is essential for genetic studies. In common
diseases such as psoriasis, the patients' general practitioners are the
best resource to confirm the diagnosis of patients. We used a letter sent
to patients' practitioners to confirm psoriasis diagnosis in a nation-wide
genetic study in France. Ethical and financial considerations, the quality
of the letter sent, and some physicians' characteristics seem to influence
the response rate to this letter.
Finally, more than 80% (70% with letters to physicians) of the psoriasis
diagnoses were confirmed, using other means to phenotype the diseases
to complete the phenotype study: in three families, a member of the family
was a physician who could confirm the diagnosis; for three other families,
from the same county (Brittany), a dermatologist from Généthon
consulted the members of the families; then, for the last cases, a dermatologist
from Généthon telephoned physicians who did not reply to
the letters.
The use of letters to physicians as a means of phenotyping patients'
diseases, after evaluation of the semiological aspects of the disease
by phone calls to the patients, is an economical, time-efficient, and
a medically acceptable compromise method of phenotyping diseases and should
be used more frequently in large genetic studies: during the 3.5 years
of the phenotype study, we evaluated 108 families, which represented 4,203
individuals, and 762 cases of putative psoriasis. We confirmed 80% of
psoriasis cases (70% with letters sent to patients' referent physicians).
Acknowledgements
We thank all the family members who participated in this study. We also
thank APLCP (Association pour la lutte contre le psoriasis) and METROBUS
for help during the study, and Susan Cure for help in writing the manuscript
in English. This study was supported by AFM (Association française
contre les myopathies) and Généthon.
Article accepted on 11/9/01
REFERENCES
1. Bhalerao J, Bowcock AM. The genetics of psoriasis: a complex
disorder of the skin and immune system. Hum Mol Genet. 1998; 7:
1537-45.
2. Ott J. Computer-simulation methods in human linkage analysis.
Proc Natl Acad Sci USA 1989; 86: 4175-8.
3. Rosenwald l'annuaire des médecins. 112th edition.
Paris: Rosenwald, 1998.
4. Dictionnaire des communes. France métropolitaine,
départements et territoires d'Outre-Mer. Panazol: Editions
Lavauzelle, 1992.
5. McAvoy BR, Kaner EFS. General practice postal surveys: a questionnaire
too far? Br Med J 1996; 313: 732-3.
6. Cartwright A. Professionals as responders: variations in and
effects of response rates to questionnaires, 1961-1977. Br Med J
1978; ii: 292-4.
7. McPherson I, Bisset A. Not another questionnaire!: eliciting
the views of general practitioners. Fam Practice 1995; 12: 335-8.
8. Carlisle R, Johnstone S. Factors influencing the response
to advertisements for general practice vacancies. Br Med J 1996;
313: 468-71.
9. Medical students face average overdraft of £1,000. Br
Med J 1998; 316: 1540.
10. Ness RB, Ukoli F, Hunt S, et al. Salary equity among
male and female internists in Pennsylvania. Ann Intern Med 2000;
133: 104-10.
11. Meier DE, Emmons CA, Wallenstein S, Quill T, Morrison RS,
Cassel CK. A national survey of physician-assisted suicide and euthanasia
in the United States. N Engl J Med 1998; 338: 1193-201.
12. Schneider EC, Epstein AM. Influence of cardiac-surgery performance
reports on referal practices and access to care. A survey of cardiovascular
specialists. N Engl J Med 1996; 335: 251-6.
13. Loi Huriet-Sérusclat n° 88-1138 du 20 décembre
1988 relative à la protection des personnes qui se prêtent
à des recherches biomédicales et loi n° 94-630 du 25
juillet 1994, relative à la protection des personnes se prêtant
à des recherches biomédicales. JO République Française
26 juillet 1994: 10747-9.
14. Rushton JL, Clark SJ, Freed GL. Pediatrician and family physician
prescription of selective serotonin reuptake inhibitors. Pediatrics
2000; 105: e82.
15. Rea JN, Newhouse ML, Halil T. Skin disease in Lambeth. A
community study of prevalence and use of medical care. Br J Prev Soc
Med 1976; 30: 107-14.
|
Version imprimable
