Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Functional Digestive Symptoms and Quality of Life in Patients with Ehlers-Danlos Syndromes: Results of a National Cohort Study on 134 Patients

  • Jean-David Zeitoun ,

    Affiliations Department of Gastroenterology and Nutrition, Saint-Antoine Hospital, APHP, Paris, France, Department of Proctology, Deaconesses Hospital, Paris, France

  • Jérémie H. Lefèvre,

    Affiliations Department of Digestive and General Surgery, Saint-Antoine Hospital, APHP, Paris, France, University Paris VI, Paris, France

  • Vincent de Parades,

    Affiliation Department of Proctology, Saint-Joseph Hospital, Paris, France

  • César Séjourné,

    Affiliation Private Office, Marolles en Hurepoix, France

  • Iradj Sobhani,

    Affiliation Department of Gastroenterology, Henri Mondor Hospital, APHP, Créteil, France

  • Benoît Coffin,

    Affiliations Department of Gastroenterology, Louis Mourier Hospital, APHP, Colombes, France, University Denis Diderot-Paris VII, Paris, France

  • Claude Hamonet

    Affiliations Department of Physical and Rehabilitation Medicine, Hôtel Dieu Hospital, APHP, Paris, France, Department of Medicine, University Paris-East-Créteil (UPEC), France


Background and Objectives

Ehlers-Danlos syndromes (EDS) are a heterogeneous group of heritable connective tissue disorders. Gastrointestinal manifestations in EDS have been described but their frequency, nature and impact are poorly known. We aimed to assess digestive features in a national cohort of EDS patients.


A questionnaire has been sent to 212 EDS patients through the French patient support group, all of which had been formally diagnosed according to the Villefranche criteria. The questionnaire included questions about digestive functional symptoms, the GIQLI (Gastrointestinal Quality of Life Index), KESS scoring system and the Rome III criteria.


Overall, 135 patients (64% response rate) completed the questionnaire and 134 were analyzable (123 women; 91%). Mean age and Body Mass Index were respectively 35±14.7 years and 24.3±6.1 kg/m2. The most common EDS subtype was hypermobility form (n=108; 80.6%). GIQLI and KESS median values were respectively 63.5 (27-117) and 19 [13.5-22]. Eighty four percent of patients had functional bowel disorders (FBD) according to the Rome III criteria. An irritable bowel syndrome according to the same criteria was observed in 64 patients (48%) and 48 patients (36%) reported functional constipation. A gastro-esophageal reflux disease (GERD) was reported in 90 patients (68.7%), significantly associated with a poorer GIQLI (60.5±16.8 versus 75.9±20.3; p<0.0001). GIQLI was also negatively impacted by the presence of an irritable bowel syndrome or functional constipation (p=0.007). There was a significant correlation between FBD and GERD.


Natural frequency of gastrointestinal manifestations in EDS seems higher than previously assessed. FBD and GERD are very common in our study population, the largest ever published until now. Their impact is herein shown to be important. A systematic clinical assessment of digestive features should be recommended in EDS.


Ehlers-Danlos syndromes (EDS) are a heterogeneous group of heritable connective tissue disorders mainly characterized by joint hypermobility, skin hyperextensibility and tissue fragility [1]. It was described at the beginning of the 20th century by two dermatologists, Edvard Ehlers and Henri-Alexandre Danlos [2,3].

In 1997, the Villefranche classification defined the 6 major forms of EDS [4]. The classical type (I), hypermobility type (II) and vascular type (III) are the most frequent clinical presentations while the 3 remaining forms (kyphoscoliosis type, arthrochalasia type, dermatosparaxis type) seem to be very rare (see Table 1). Mutations in type V and type III collagen are respectively involved in classical and vascular EDS while the 3 last forms of EDS are biologically related to the processing of type I collagen. However, genetic background and pathophysiology of EDS remain to be fully elucidated. Numerous articles in the literature have focused on the vascular type because it is associated with the poorer prognosis [5] but it appears that this body of publications is out of proportion compared to the natural prevalence of respective forms. Hypermobility type is probably more frequent but might be very often undiagnosed [6]. Although gastrointestinal manifestations have already been reported in EDS [7], mainly hemorrhage or perforation, little attention has been paid to functional non-life-threatening gastro-intestinal symptoms which could be frequent considering the importance of collagens in the gut. It has to be noted that most studies recruited a small number of patients. The aim of this study was to describe the type and the frequency of gastrointestinal functional disorders and their impact on quality of life in a large cohort of patients with EDS.

Ehlers-Danlos Syndrome   InheritanceMajor diagnostic criteria
I, classical typeAutosomal dominantSkin hyperextensibility
Widened atrophic scars
Joint hypermobility
II, hypermobility typeAutosomal dominantSkin involvement
Generalized joint hypermobility
III, vascular typeAutosomal dominantThin, translucent skin
Arterial/intestinal/uterine fragility or rupture
Extensive bruising
Characteristic facial appearance
IV, kyphoscoliosis typeAutosomal recessiveGeneralized joint laxity
Severe muscle hypotonia at birth
Scoliosis at birth, progressive
Scleral fragility and rupture of the ocular globe
V, arthrochalasia typeAutosomal dominantSevere generalized joint hypermobility with recurrent subluxations
Congenital bilateral hip disclocation
VI, dermatosparaxis typeAutosomal recessiveSevere skin fragility
Sagging, redundant skin

Table 1. Nosology from the Villefranche Classification [1].

Download CSV

Materials and Methods

An original standardized questionnaire was sent by e-mail or Internet from September 2011 to July 2012 to 212 patients affected by EDS through the French patient support group (Apprivoiser les Syndromes d’Ehlers-Danlos, ASED, All of the patients had been clinically assessed at least once by a single national expert practitioner in EDS (CH) and all were formally diagnosed as having EDS according to Villefranche criteria [1]. Joint hypermobility was assessed using the Beighton Scale [8]. There was no incentive to complete the questionnaire. Patients who did not complete the survey after one month received a reminder once. It was preplanned that patients who would have responded to the questionnaire improperly would not be recalled. The questionnaire included relevant general and demographic characteristics, and questions about digestive symptoms. Gastro-esophageal reflux disease (GERD) was defined as it is usually done in the scientific literature as the occurrence of either heartburn and (or) regurgitations. Most other esophageal and extraesophageal symptoms of GERD were included in the questionnaire according to their listing in a reference paper [9]. Additionally, 3 internationally validated or at least commonly accepted questionnaires were included which were the following: irritable bowel syndrome (IBS) and functional constipation (FC) were defined according to the Rome III criteria. Severity of constipation was determined by using the Kess score. Impact of these symptoms on quality of life was measured by the Gastrointestinal Quality of Life Index [10] (GIQLI). All these 3 questionnaires had been translated in French. For the Rome III criteria and the Kess score, we used the French versions of the French Group of Neurogastroenterology (Groupe Français de Neurogastroentérologie, GFNG) which is the official study group of motility and sensitivity, affiliated to the French National Society of Gastroenterology (SNFGE) and the European Society of Neurogastroenterology and Motility (ESNM). For the GIQLI, we used a validated French version of the questionnaire [11]. Both English and French complete versions of the questionnaires can be found in the Appendix. Data were collected in September 2012 by a single person and entered into a dedicated and anonymized database. This study was approved by the French National Commission for Data Protection (Commission Nationale Informatique et Libertés). All patients were fully informed according to the French ethics law and gave written consent.

Statistical analysis

Statistical analysis was performed by using JMP9 (SAS Institute, USA). Data are shown as the prevalence, mean (standard deviation), or median (range). Continuous data were compared by using the Mann-Whitney U test. Comparison of mean values between three groups (IBS, FC, and no IBS or FC) was performed using ANOVA test. All statistical tests were two-sided, with the threshold of significance set at p<0.05.


Out of the 212 patients at baseline, 113 patients returned a questionnaire by e-mail and 22 by mail. Overall, 135 returned a completed questionnaire (64% response rate) among which 134 were analyzable. Demographics and clinical characteristics are shown in Table 2. Gastrointestinal manifestations had been starting before other symptoms of EDS in 54 patients (44.6%) and before EDS formal diagnosis in 96 patients (74.4%). First manifestation of EDS was observed before the age of 18 in 94 patients (69.6%).

Characteristicsn (%)
Age – mean (± SD)35 (14.7)
Female 123 (91)
Weight – mean (kg ± SD)66.6 ± 18.9
Height – mean (cm ± SD)164.9 ± 10.2
Body mass index – mean (± SD)24.3 ± 6.1
Age at first symptoms of EDS
Infant (0-2 years)26 (19.3%)
Childhood 2-10 years)41 (30.4%)
Adolescence (11-17 years)27 (20%)
Adult (> 18 years)29 (21.5%)
Unknown12 (8.9%)
Tobacco 39 (31.0)
Daily work 39 (32.5)
Classical type 11 (8.2)
Hypermobility type 108 (80.6)
Vascular type 3 (2.2)
Other – Mixed forms12 (9)

Table 2. Demographics and clinical characteristics.

SD = Standard Deviation.
Download CSV

Upper GI symptoms: Gastro-esophageal reflux disease (GERD) and dyspepsia

Frequencies of different symptoms of GERD or dyspepsia are shown in Table 3. One hundred and seven patients complained from heartburn and/or regurgitations (79.3%). Seventy-two patients (55%) had undergone upper endoscopy, among which 33 of them (45.8%) were declared to be normal or unremarkable. Nineteen patients had a hiatal hernia on endoscopy.

Symptomsn (%)
Heartburn90 (68.7)
Regurgitations90 (68.7)
Symptoms worsened by decubitus82 (62.6)
Chronic cough47 (36.2)
Laryngitis75 (56.8)
Erosion of dental enamel67 (51.5)
Asthma58 (45)
Dysphagia82 (62.6)
Epigastric pain104 (78.8)
Nausea92 (70.8)
Postprandial fullness88 (67.2)
Belching91 (70.5)

Table 3. Frequencies of upper GI symptoms: gastro-esophageal reflux disease (GERD) or dyspepsia.

Download CSV

Lower GI symptoms: irritable bowel syndrome and functional constipation

Sixty-four patients (48%) reported irritable bowel syndrome (IBS) according to the Rome III criteria (Table 4) with a homogenous repartition according to stool consistency. Forty-eight patients (36%) displayed functional constipation (FC) according to the Rome III criteria. KESS median score was 19 [13.5-22] and 117 patients (87.3%) had a KESS score > 9, which is the generally accepted cut-off for constipation. Patients with functional constipation had a significant worse KESS score (21.5±0.82 vs. 15.8±0.84; p<0.0001). There was a statistically significant association between FBD and GERD: 89.2% (91/102) patients with FBD also had GERD whereas 59.1% (13/22) of patients without FBD were displaying GERD symptoms (p=0.002) (Figure 1).

SymptomsN (%)Diagnosis
Recurrent abdominal pain or discomfort at least 3 days per month in the last three months associated with:
Improvement with defecation 78
Onset associated with a change in frequency of stool 55
Onset associated with a change in form (appearance) of stool 66
Irritable Bowel Syndrome (IBS) diagnosis
3 associated above features4164 patients having IBS according to the Rome III criteria
2 associated above features23
1 above feature38
IBS with constipation18 (28%)
IBS with diarrhea18 (28%)
IBS mixed 19 (30%)
Unsubtyped IBS9 (14%)
Diagnostic criteria for functional constipation
Fewer than 3 defecations per week 37
Lumpy or hard stools in at least 25% of defecations 64
Straining during at least 25% of defecations 93
Sensation of incomplete evacuation for at least 25% of defecations 98
Sensation of anorectal obstruction/blockage for at least 25% of defecations54
Manual maneuvers to facilitate at least 25% of defecations 29
Functional Constipation Diagnosis
6 associated above features548 patients having functional constipation according to the Rome III criteria
5 associated above features8
4 associated above features6
3 associated above features16
2 associated above features13

Table 4. Rome III criteria for Irritable Bowel Syndrome (IBS) and Functional Constipation in the Ehlers-Danlos Syndromes cohort.

Download CSV
Figure 1. Flow Chart of the functional bowel disorders among the population of Ehlers-Danlos syndrome.

Impact on quality of life

Median GIQLI was 63.5 [27-117]. The mean score was significantly lower when compared with a French control population of 238 individuals [11] (65.3±1.65 vs. 128.0±0.81; p<0.0001; see Figure 2). Details and subscales results are presented in Table 5. All subscales scores were significantly lower for patients with EDS when compared to the control groups (p<0.0001), see Figure 3.

Figure 2. GIQLI score (individual spot and means) according to the presence of Ehlers-Danlos syndrome.

ItemsMean±SDSubscale score
Theoretical RangeMean ± SDMedian (min-max)
Physical Well-Being0-4011.75±5.5911(1-32)
Item 8 (Pleasure and appetite)1.60±1.18
Item 15 (Fatigue)0.46±0.71
Item 16 (Feeling unwell)1.13±0.92
Item 18 (Appearance)1.62±1.47
Item 19 (Physical strength)0.82±1.13
Item 20 (Endurance)0.35±0.77
Item 21 (Feeling unfit)0.40±0.78
Item 22 (Daily activities)1.94±1.12
Item 23 (Leisure activities)1.44±1.12
Item 33 (Nausea)2.08±1.22
Gastrointestinal digestion 0-4017.96±7.7917 (1-38)
Item 1 (Pain in abdomen)1.65±0.97
Item 2 (Fullness in abdomen)1.63±1.27
Item 3 (Bloating)1.47±1.17
Item 4 (Flatus)1.89±1.28
Item 5 (Burping/belching)1.94±1.24
Item 6 (Abdominal noises)1.73±1.18
Item 27 (Regurgitation)2.01±1.22
Item 28 (Eating speed)2.06±1.38
Item 32 (Constipation)1.55±1.32
Item 35 (Heartburn)2.09±1.28
Gastrointestinal defecation0-2416.01±4.3116 (3-24)
Item 7 (Bowel frequency)2.56±1.26
Item 26 (Impaired sexual life)1.55±1.26
Item 30 (Bowel Urgency)2.44±1.18
Item 31 (Diarrhoea)2.82±1.12
Item 34 (Blood in stool)3.45±0.94
Item 36 (uncontrolled stools)3.47±0.91
Mental Well Being Scale0-209.80±3.9810 (0-19)
Item 10 (Coping with stress)1.90±1.02
Item 11 (Sad about illness)2.02±1.07
Item 12 (Nervous about illness)2.13±1.20
Item 13 (Happy with life)2.23±0.96
Item 14 (Frustrated by illness)1.56±0.92
Items not included in a subscale
Item 9 (Restricted eating) 2.10±1.38
Item 17 (Wake up at night)0.89±1.27
Item 24 (Bothered by treatments)2.64±1.19
Item 25 (Worsened relations)1.60±1.26
Item 29 (Dysphagia)2.61±1.19
Overall score0-14465.3±19.163.5 (27-117)

Table 5. Details of GIQLI score and subscales.

SD = Standard Deviation; GIQLI = Gastrointestinal Quality of Life Index.
Download CSV
Figure 3. Subscales of the GIQLI between patients with EDS (box with white lines) and control population (filled box) and results of the Student comparison.

GERD was significantly associated with a worst GIQLI (60.5 ± 16.8 versus 75.9 ± 20.3; p<0.0001). GIQLI was also significantly lower in case of IBS or FC compared to no FBD (p=0.0007) (Figure 4).

Figure 4. GIQLI score according to the presence and the type of functional bowel disorders (FC: functional constipation; IBS: irritable bowel syndrome).


In the present survey, we have shown that the frequency of functional gastrointestinal manifestations in EDS was highly prevalent, much higher than previously assessed. IBS, functional constipation and GERD were present in respectively 48%, 36% and 79% of our study population, the largest ever published until now. Their impact is herein shown to be important.

Ehlers-Danlos syndromes (EDS) are a genetically and clinically heterogeneous group of disorders characterized by a fragility of the soft connective tissues [12]. The Villefranche classification recognizes six EDS subtypes among which classical, hypermobility and vascular types are the most frequent ones [1]. Most publications concerning gut symptoms in patients with EDS are dealing with vascular type, certainly because it is the most serious one, but gut symptoms occurring in classical or hypermobility types, which affect the majority of patients are paradoxically less documented. No standardized approach has ever been adopted to describe them properly. Only life-threatening complications such as spontaneous perforation or massive bleeding have been reported but one might question the representativeness of such reports regarding the whole population of EDS since it was generally in vascular types of EDS. Our survey is, to our knowledge, the largest in an EDS national cohort assessing gastrointestinal profile of affected patients. All subjects have been formally diagnosed after clinical examination by a single national expert, (CH) according to the validated international Villefranche criteria. Thus, diagnosis of EDS in the respondents of our survey is thought to be reliable and other phenotypically related conditions must have been excluded. Another strength of our study is the use of internationally validated questionnaire and scoring systems to assess clinical features of recruited patients. The major findings of the current study are that gastrointestinal manifestations are very common and generally not specific, frequently important and that they can have a strong impact on quality of life. Eighty four percent of the studied population had FBD among which 57% had IBS and 43% functional constipation. Approximately 80% of patients had GERD. There was a statistically significant association between upper and lower GI symptoms. Overall, median GIQLI was 63.5 [51.8-76.8], which is extremely low compared to most publications and it is of note that GERD and lower GI symptoms negatively influenced this scoring system. All these findings clearly indicate that digestive manifestations in EDS are of major relevance and may have been previously underdiagnosed and undertreated. All reported symptoms are remarkably nonspecific and this could partly explain why little attention has been paid until now to these clinical manifestations. In addition, no severe complication has been described in our population, emphasizing the clear distinction between rare but serious complications of EDS vascular type and other common benign but disabling manifestations, for which literature is scarce. Many questions remain unanswered, among which the pathophysiology of reported symptoms. One might hypothesize that tissue hyperextensibility of the gastrointestinal tract could play a role but proprioceptive disorders as well as dysautonomic syndrome which are very common in EDS [13] could also contribute to gastrointestinal manifestations. In addition, the marked preponderance of affected women vs. men in EDS and especially hypermobile type, although previously documented, is still an unexplained feature [6]. Also, current treatment for gastrointestinal symptoms is empirical, often unsatisfactory (data not shown) and further research is needed. At least, a systematic assessment of gastrointestinal symptoms should be recommended in EDS patients in order not to miss a major source of complaints in this population. Whether endoscopic examinations are necessary and at high risk of complication, in particular perforation, is an unresolved question. In our experience, upper gastrointestinal endoscopy seems safe and useful to detect Barrett’s esophagus in this population with a high rate of GERD. On the other hand, the matter of colonoscopy is more sensitive. Indeed, the risk of perforation is clearly significant in vascular EDS and colonoscopy should be strongly discouraged in this population. Additionally, the risk of any complication (perforation or bleeding) is theoretically increased in other EDS subtypes, although not quantifiable and although no patient in our cohort underwent any complication. CT colonography could be an alternative option to rule out polyps or tumor but one should keep in mind the risk of repeated irradiating exams in this population with frequent orthopedic complications.

Our study has several limitations. The first one is the absence of control group (except regarding GIQLI for which we used a historical control French population) which precludes definitive conclusions from this survey. However, there is an abundant literature about FBD and GERD and comparisons with reported natural frequencies and severity in historical populations allow us to assume that gastrointestinal manifestations in EDS are particularly common and linked to the condition. For instance, a French survey conducted on a nationally representative sample reported a prevalence of IBS (according to the Rome II criteria which were slightly different from the Rome III criteria) to be lower than 5% [14]. Another epidemiologic mail survey found that prevalence of GERD reached 8% in another sample of 8000 subjects representative of the French adult population [15]. In addition, we cannot rule out the possibility of response bias, although our percentage of returned questionnaires was slightly higher than the generally required 60% response rate [16]. Although many patients of the questioned cohort had genetic testing, we were not allowed to collect these data in the present study, which is another potential deficiency. However, most of our population was affected by hypermobility type, consistently with literature and the genetic basis of this EDS form is lacking. At last, we cannot establish how many patients with differential diagnosis (such as neuromuscular disorders or other connective tissue disorders) were excluded from the present cohort.

In summary, our study is to date the largest conducted survey specifically assessing the natural frequency, nature and impact of functional gastrointestinal manifestations in EDS. It emerges that digestive manifestations are extremely common, most frequently nonspecific and not serious but with major consequences on quality of life. A systematic clinical assessment should be recommended in EDS population and further studies are needed to elucidate the pathophysiology of these disorders and to improve therapeutic management.

Supporting Information


All authors are grateful to patients who responded to the submitted questionnaire and would like to thank the French patient support group for their help in collecting data. JD Zeitoun would like to thank Pr Hervé Maisonneuve for his precious advice and Pr Karem Slim for his gracious help by giving the data of his pivotal study.

Author Contributions

Contributed to the conception of the study: JDZ JHL VdP CS IS BC CH. Contributed to the acquisition of data and their interpretation: JDZ JHL VdP CS IS BC CH. Contributed to the final draft of the article or to its critical revision: JDZ JHL VdP CS IS BC CH. Approved the final version: JDZ JHL VdP CS IS BC CH. Proceeded to clinical examination for all included patients: CH. Performed the main part of the statistical analysis: JHL.


  1. 1. Beighton P, De Paepe A, Steinmann B, Tsipouras P, Wenstrup RJ (1998) Ehlers-Danlos syndromes: revised nosology, Villefranche, 1997. Ehlers-Danlos National Foundation (USA) and Ehlers-Danlos Support Group (UK). Am J Med Genet 77: 31-37.
  2. 2. Danlos M (1908) Un cas de cutis laxa avec tumeurs par contusion chronique des coudes et des genoux (xanthome juvénile pseudo-diabétique de MM Hallopeau et Marc de Lépinay). Bull Soc Franc Derm Syph 19: 70-72.
  3. 3. Ehlers E (1901) Cutis laxa, neigung zu haemorrhagien in der haut, lockerung mehrerer artikulationen. Derm Zschr 18: 173-175.
  4. 4. De Paepe A, Malfait F (2012) The Ehlers-Danlos syndrome, a disorder with many faces. Clin Genet 82: 1-11. doi: PubMed: 22353005.
  5. 5. Pepin M, Schwarze U, Superti-Furga A, Byers PH (2000) Clinical and genetic features of Ehlers-Danlos syndrome type IV, the vascular type. N Engl J Med 342: 673-680. doi: PubMed: 10706896.
  6. 6. Castori M, Camerota F, Celletti C, Danese C, Santilli V et al. (2010) Natural history and manifestations of the hypermobility type Ehlers-Danlos syndrome: a pilot study on 21 patients. Am J Med Genet A 152A: 556-564. doi: PubMed: 20140961.
  7. 7. Beighton PH, Murdoch JL, Votteler T (1969) Gastrointestinal complications of the Ehlers-Danlos syndrome. Gut 10: 1004-1008. doi: PubMed: 5308459.
  8. 8. Beighton P, Solomon L, Soskolne CL (1973) Articular mobility in an African population. Ann Rheum Dis 32: 413-418. doi: PubMed: 4751776.
  9. 9. Kahrilas PJ (2008) Clinical practice. Gastroesophageal reflux disease. N Engl J Med 359: 1700-1707. doi: PubMed: 18923172.
  10. 10. Eypasch E, Williams JI, Wood-Dauphinee S, Ure BM, Schmülling C et al. (1995) Gastrointestinal Quality of Life Index: development, validation and application of a new instrument. Br J Surg 82: 216-222. doi: PubMed: 7749697.
  11. 11. Slim K, Bousquet J, Kwiatkowski F, Lescure G, Pezet D et al. (1999) First validation of the French version of the Gastrointestinal Quality of Life Index (GIQLI). Gastroenterol Clin Biol 23: 25-31. PubMed: 10219601.
  12. 12. Callewaert B, Malfait F, Loeys B, De Paepe A (2008) Ehlers-Danlos syndromes and Marfan syndrome. Best Pract Res Clin Rheumatol 22: 165-189. doi: PubMed: 18328988.
  13. 13. Castori M, Morlino S, Celletti C, Celli M, Morrone A et al. (2012) Management of pain and fatigue in the joint hypermobility syndrome (a.k.a. Ehlers-Danlos syndrome, hypermobility type): principles and proposal for a multidisciplinary approach. Am J Med Genet A 158A: 2055-2070. doi: PubMed: 22786715.
  14. 14. Dapoigny M, Bellanger J, Bonaz B, Bruley des Varannes S, Bueno L et al. (2004) Irritable bowel syndrome in France: a common, debilitating and costly disorder. Eur J Gastroenterol Hepatol 16: 995-1001. doi: PubMed: 15371923.
  15. 15. Bretagne JF, Richard-Molard B, Honnorat C, Caekaert A, Barthélemy P (2006) Gastroesophageal reflux in the French general population: national survey of 8000 adults. Presse Med 35: 23-31. doi: PubMed: 16462660.
  16. 16. Johnson TP, Wislar JS (2012) Response rates and nonresponse errors in surveys. JAMA 307: 1805-1806. doi: PubMed: 22550194.