Extent of non-publication in cohorts of studies approved by research ethics committees or included in trial registries

BACKGROUND: The synthesis of published research in systematic reviews is essential when providing evidence to inform clinical and health policy decision-making. However, the validity of systematic reviews is threatened if journal publications represent a biased selection of all studies that have been conducted (dissemination bias). To investigate the extent of dissemination bias we conducted a systematic review that determined the proportion of studies published as peer-reviewed journal articles and investigated factors associated with full publication in cohorts of studies (i) approved by research ethics committees (RECs) or (ii) included in trial registries. METHODS AND FINDINGS: Four bibliographic databases were searched for methodological research projects (MRPs) without limitations for publication year, language or study location. The searches were supplemented by handsearching the references of included MRPs. We estimated the proportion of studies published using prediction intervals (PI) and a random effects meta-analysis. Pooled odds ratios (OR) were used to express associations between study characteristics and journal publication. Seventeen MRPs (23 publications) evaluated cohorts of studies approved by RECs; the proportion of published studies had a PI between 22% and 72% and the weighted pooled proportion when combining estimates would be 46.2% (95% CI 40.2%-52.4%, I2 = 94.4%). Twenty-two MRPs (22 publications) evaluated cohorts of studies included in trial registries; the PI of the proportion published ranged from 13% to 90% and the weighted pooled proportion would be 54.2% (95% CI 42.0%-65.9%, I2 = 98.9%). REC-approved studies with statistically significant results (compared with those without statistically significant results) were more likely to be published (pooled OR 2.8; 95% CI 2.2-3.5). Phase-III trials were also more likely to be published than phase II trials (pooled OR 2.0; 95% CI 1.6-2.5). The probability of publication within two years after study completion ranged from 7% to 30%. CONCLUSIONS: A substantial part of the studies approved by RECs or included in trial registries remains unpublished. Due to the large heterogeneity a prediction of the publication probability for a future study is very uncertain. Non-publication of research is not a random process, e.g., it is associated with the direction of study findings. Our findings suggest that the dissemination of research findings is biased.

Evaluation of prenatal diagnosis of cleft lip with or without cleft palate and cleft palate by ultrasound: experience from 20 European registries. EUROSCAN study group.

Ultrasound scans in the mid-trimester of pregnancy are now a routine part of antenatal care in most European countries. Using data from registries of congenital anomalies a study was undertaken in Europe. The objective of the study was to evaluate prenatal detection of cleft lip with or without cleft palate (CL(P)) and cleft palate (CP). All CL(P) and CPs suspected prenatally and identified at birth in the period 1996-98 were registered from 20 Congenital Malformation Registers from the following European countries: Austria, Croatia, Denmark, France, Germany, Italy, Lithuania, Spain, Switzerland, The Netherlands, UK, Ukraine. These registries followed the same methodology. A total of 709,027 births were covered; 7758 cases with congenital malformations were registered. Included in the study were 751 cases reported with facial clefts: 553 CL(P) and 198 CP. The prenatal diagnosis by transabdominal ultrasound of CL(P) was made in 65/366 cases with an isolated malformation, in 32/62 cases with chromosomal anomaly, in 30/89 cases with multiple malformations and in 21/36 syndromic cases. The prenatal diagnosis of CP was made in 13/198 cases. One hundred pregnancies were terminated (13%); in 97 of these the cleft was associated with other malformations.

Evaluation of prenatal ultrasound diagnosis of fetal abdominal wall defects by 19 European registries.

OBJECTIVES: To evaluate the current effectiveness of routine prenatal ultrasound screening in detecting gastroschisis and omphalocele in Europe. DESIGN: Data were collected by 19 congenital malformation registries from 11 European countries. The registries used the same epidemiological methodology and registration system. The study period was 30 months (July 1st 1996-December 31st 1998) and the total number of monitored pregnancies was 690,123. RESULTS: The sensitivity of antenatal ultrasound examination in detecting omphalocele was 75% (103/137). The mean gestational age at the first detection of an anomaly was 18 +/- 6.0 gestational weeks. The overall prenatal detection rate for gastroschisis was 83% (88/106) and the mean gestational age at diagnosis was 20 +/- 7.0 gestational weeks. Detection rates varied between registries from 25 to 100% for omphalocele and from 18 to 100% for gastroschisis. Of the 137 cases of omphalocele less than half of the cases were live births (n = 56; 41%). A high number of cases resulted in fetal deaths (n = 30; 22%) and termination of pregnancy (n = 51; 37%). Of the 106 cases of gastroschisis there were 62 (59%) live births, 13 (12%) ended with intrauterine fetal death and 31 (29%) had the pregnancies terminated. CONCLUSIONS: There is significant regional variation in detection rates in Europe reflecting different policies, equipment and the operators' experience. A high proportion of abdominal wall defects is associated with concurrent malformations, syndromes or chromosomal abnormalities, stressing the need for the introduction of repeated detailed ultrasound examination as a standard procedure. There is still a relatively high rate of elective termination of pregnancies for both defects, even in isolated cases which generally have a good prognosis after surgical repair.

Paper 3: EUROCAT data quality indicators for population-based registries of congenital anomalies.

The European Surveillance of Congenital Anomalies (EUROCAT) network of population-based congenital anomaly registries is an important source of epidemiologic information on congenital anomalies in Europe covering live births, fetal deaths from 20 weeks gestation, and terminations of pregnancy for fetal anomaly. EUROCAT's policy is to strive for high-quality data, while ensuring consistency and transparency across all member registries. A set of 30 data quality indicators (DQIs) was developed to assess five key elements of data quality: completeness of case ascertainment, accuracy of diagnosis, completeness of information on EUROCAT variables, timeliness of data transmission, and availability of population denominator information. This article describes each of the individual DQIs and presents the output for each registry as well as the EUROCAT (unweighted) average, for 29 full member registries for 2004-2008. This information is also available on the EUROCAT website for previous years. The EUROCAT DQIs allow registries to evaluate their performance in relation to other registries and allows appropriate interpretations to be made of the data collected. The DQIs provide direction for improving data collection and ascertainment, and they allow annual assessment for monitoring continuous improvement. The DQI are constantly reviewed and refined to best document registry procedures and processes regarding data collection, to ensure appropriateness of DQI, and to ensure transparency so that the data collected can make a substantial and useful contribution to epidemiologic research on congenital anomalies.

Comparative effectiveness research across two spine registries.

BACKGROUND: Comparative effectiveness research in spine surgery is still a rarity. In this study, pain alleviation and quality of life (QoL) improvement after lumbar total disc arthroplasty (TDA) and anterior lumbar interbody fusion (ALIF) were anonymously compared by surgeon and implant. METHODS: A total of 534 monosegmental TDAs from the SWISSspine registry were analyzed. Mean age was 42 years (19-65 years), 59% were females. Fifty cases with ALIF were documented in the international Spine Tango registry and used as concurrent comparator group for the pain analysis. Mean age was 46 years (21-69 years), 78% were females. The average follow-up time in both samples was 1 year. Comparison of back/leg pain alleviation and QoL improvement was performed. Unadjusted and adjusted probabilities for achievement of minimum clinically relevant improvements of 18 VAS points or 0.25 EQ-5D points were calculated for each surgeon. RESULTS: Mean preoperative back pain decreased from 69 to 30 points at 1 year (ØΔ 39pts) after TDA, and from 66 to 27 points after ALIF (ØΔ 39pts). Mean preoperative QoL improved from 0.34 to 0.74 points at 1 year (ØΔ 0.40pts). There were surgeons with better patient selection, indicated by lower adjusted probabilities reflecting worsening of outcomes if they had treated an average patient sample. ALIF had similar pain alleviation than TDA. CONCLUSIONS: Pain alleviation after TDA and ALIF was similar. Differences in surgeon's patient selection based on pain and QoL were revealed. Some surgeons seem to miss the full therapeutic potential of TDA by selecting patients with lower symptom severity.

High constant incidence rates of second primary cancers of the head and neck: a pooled analysis of 13 cancer registries.

Scanty data are available on the incidence (i.e., the absolute risk) of second cancers of the head and neck (HN) and its pattern with age. We investigated this issue using data from a multicentric study of 13 population-based cancer registries from Europe, Canada, Australia and Singapore for the years 1943-2000. A total of 99,257 patients had a first primary HN cancer (15,985 tongue, 22,378 mouth, 20,758 pharyngeal, and 40,190 laryngeal cancer), contributing to 489,855 person-years of follow-up. A total of 1,294 of the patients (1.3%) were diagnosed with second HN cancers (342 tongue, 345 mouth, 418 pharynx and 189 larynx). Male incidence rates of first HN cancer steeply increased from 0.68/100,000 at age 30-34 to 46.2/100,000 at age 70-74, and leveled off at older age; female incidence increased from 0.50/100,000 at age 30-34 to 16.5/100,000 at age 80-84. However, age-specific incidence of second HN cancers after a first HN cancer in men was around 200-300/100,000 between age 40-44 and age 70-74 and tended to decline at subsequent ages (150/100,000 at age 80-84); in women, incidence of second HN cancers was around 200-300/100,000 between age 45-49 and 80-84. The patterns of age-specific incidence were consistent for different subsites of second HN cancer and sexes; moreover, they were similar for age-specific incidence of first primary HN cancer in patients who subsequently developed a second HN cancer. The incidence of second HN cancers does not increase with age, but remains constant, or if anything, decreases with advancing age.

Turner syndrome: evaluation of prenatal diagnosis in 19 European registries.

This study evaluated the prenatal diagnosis of Turner syndrome by ultrasound examination in an unselected population from all over Europe. Data from 19 congenital malformation registries from 11 European countries were analyzed. Turner syndrome was diagnosed in 125 cases (7.2%) in a total of 1,738 chromosome abnormalities. Sixty-seven percent of cases were detected prenatally by ultrasound examination due to the presence of congenital defects. The most frequent anomalies were cystic hygroma (59.5%) and hydrops fetalis (19%). The most frequent karyotype was 45,X (81.6%) followed by different types of mosaicism (16.8%). Significant differences in congenital defects (P = 0.0003) were observed between 45,X karyotypes and 45,X mosaicism cases. Prenatal counseling for 45,X mosaicism should take into account the expectation of a milder phenotype. In 78.6% of cases diagnosed by ultrasound examination due to congenital anomalies, the pregnancy was terminated. Prenatal detection of Turner syndrome by ultrasound examination was high in this unselected population.

Prenatal ultrasonographic detection of gastrointestinal obstruction: results from 18 European congenital anomaly registries.

OBJECTIVES: We evaluated the prenatal detection of gastrointestinal obstruction (GIO, including atresia, stenosis, absence or fistula) by routine ultrasonographic examination in an unselected population all over Europe. METHODS: Data from 18 congenital malformation registries in 11 European countries were analysed. These multisource registries used the same methodology. All fetuses/neonates with GIO confirmed within 1 week after birth who had prenatal sonography and were born during the study period (1 July 1996 to 31 December 1998) were included. RESULTS: There were 670 793 births in the area covered and 349 fetuses/neonates had GIO. The prenatal detection rate of GIO was 34%; of these 40% were detected < or = 24 weeks of gestation (WG). A total of 31% (60/192) of the isolated GIO were detected prenatally, as were 38% (59/157) of the associated GIO (p=0.26). The detection rate was 25% for esophageal obstruction (31/122), 52% for duodenal obstruction (33/64), 40% for small intestine obstruction (27/68) and 29% for large intestine obstruction (28/95) (p=0.002). The detection rate was higher in countries with a policy of routine obstetric ultrasound. Fifteen percent of pregnancies were terminated (51/349). Eleven of these had chromosomal anomalies, 31 multiple malformations, eight non-chromosomal recognized syndromes, and one isolated GIO. The participating registries reflect the various national policies for termination of pregnancy (TOP), but TOPs after 24 WG (11/51) do not appear to be performed more frequently in countries with a liberal TOP policy. CONCLUSION: This European study shows that the detection rate of GIO depends on the screening policy and on the sonographic detectability of GIO subgroups.

Second primary cancers in the Vaud and Neuchâtel Cancer Registries

An increasing proportion of new cancers is registered in patients who have received a previous cancer diagnosis. As data are inconsistent across studies, we provided information for populations long covered by valid cancer registration. Data were derived from the Swiss cancer Registries of Vaud and Neuchâtel (885 000 inhabitants). Patients diagnosed with a new malignancy (except skin basal and squamous cell carcinomas) during the period 2005-2010 were included. Over the period 2005-2010, 24 859 patients were registered with incident cancer. Of these, 3127 (13%) had multiple primary cancers and 578 (2.3%) were synchronous. Breast, prostate, colorectum, skin, melanomas, and squamous cell carcinomas of the head and neck (SHN) and bladder/ureter were the most common sites of first neoplasms, whereas breast, lung, colorectum, prostate, melanoma, and SHN were the most common sites of second neoplasms. The most common pairing was breast with breast (31% synchronous), followed by the bladder/ureter with the prostate (72% synchronous), prostate with the colorectum, SHN with SHN, and SHN with lung. Five-year crude survival of patients with synchronous cancers (34%) was not significantly lower than that of patients with single neoplasms (39%).

Registries in rheumatological and musculoskeletal conditions. Paediatric Behçet's disease: an international cohort study of 110 patients. One-year follow-up data.

OBJECTIVE: To set-up an international cohort of patients suspected with Behçet's disease (BD). The cohort is aimed at defining an algorithm for definition of the disease in children. METHODS: International experts have defined the inclusion criteria as follows: recurrent oral aphthosis (ROA) plus one of following-genital ulceration, erythema nodosum, folliculitis, pustulous/acneiform lesions, positive pathergy test, uveitis, venous/arterial thrombosis and family history of BD. Onset of disease is <16 years, disease duration is ≤3 years, future follow-up duration is ≥4 years and informed consent is obtained. The expert committee has classified the included patients into: definite paediatric BD (PED-BD), probable PED-BD and no PED-BD. Statistical analysis is performed to compare the three groups of patients. Centres document their patients into a single database. RESULTS: At January 2010, 110 patients (56 males/54 females) have been included. Mean age at first symptom: 8.1 years (median 8.2 years). At inclusion, 38% had only one symptom associated with ROA, 31% had two and 31% had three or more symptoms. A total of 106 first evaluations have been done. Seventeen patients underwent the first-year evaluation, and 36 had no new symptoms, 12 had one and 9 had two. Experts have examined 48 files and classified 30 as definite and 18 as probable. Twenty-six patients classified as definite fulfilled the International Study Group criteria. Seventeen patients classified as probable did not meet the international criteria. CONCLUSION: The expert committee has classified the majority of patients in the BD group although they presented with few symptoms independently of BD classification criteria.

Evaluation of prenatal diagnosis of congenital heart diseases by ultrasound: experience from 20 European registries.

OBJECTIVES: To evaluate prenatal diagnosis of congenital heart diseases by ultrasound investigation in well-defined European populations. DESIGN: Data from 20 registries of congenital malformations in 12 European countries were included. The prenatal ultrasound screening programs in the countries ranged from no routine screening to three ultrasound investigations per patient routinely performed. RESULTS: There were 2454 cases with congenital heart disease with an overall prenatal detection rate of 25%. Termination of pregnancy was performed in 293 cases (12%). There was considerable variation in prenatal detection rate between regions, with the lowest detection rates being in countries without ultrasound screening (11%) and in Eastern European countries (Croatia, Lithuania and Ukraine; 8%). In Western European countries with ultrasound screening, detection rate ranged from 19-48%. There was a significant difference in prenatal detection rate and proportion of induced abortions between isolated congenital heart disease and congenital heart disease associated with chromosome anomalies, multiple malformations and syndromes (P < 0.0001). There were 1694 cases with isolated congenital heart disease of which 16% were diagnosed prenatally. Malformations affecting the size of the ventricles were detected prenatally in half of the cases. CONCLUSIONS: Prenatal detection rate of congenital heart disease varies significantly between countries even with the same screening recommendations. The presence of associated malformations significantly increases the prenatal detection rate.

The TREAT-NMD Duchenne Muscular Dystrophy Registries: Conception, Design, and Utilization by Industry and Academia.

Duchenne muscular dystrophy (DMD) is an X-linked genetic disease, caused by the absence of the dystrophin protein. Although many novel therapies are under development for DMD, there is currently no cure and affected individuals are often confined to a wheelchair by their teens and die in their twenties/thirties. DMD is a rare disease (prevalence <5/10,000). Even the largest countries do not have enough affected patients to rigorously assess novel therapies, unravel genetic complexities, and determine patient outcomes. TREAT-NMD is a worldwide network for neuromuscular diseases that provides an infrastructure to support the delivery of promising new therapies for patients. The harmonized implementation of national and ultimately global patient registries has been central to the success of TREAT-NMD. For the DMD registries within TREAT-NMD, individual countries have chosen to collect patient information in the form of standardized patient registries to increase the overall patient population on which clinical outcomes and new technologies can be assessed. The registries comprise more than 13,500 patients from 31 different countries. Here, we describe how the TREAT-NMD national patient registries for DMD were established. We look at their continued growth and assess how successful they have been at fostering collaboration between academia, patient organizations, and industry.

La place des registres dans la surveillance des maladies transmissibles. [The role of registries in the surveillance of transmissible diseases].

Registries are among the oldest methods used in public health for epidemiological surveillance and decision making in the area of communicable diseases. Although other sources of data are now available in many developed countries, registries still provide important information. This article reviews the main aims and characteristics of modern registries, providing several examples of current epidemiological problems. Practical advantages and disadvantages of registries are also discussed, as well as some developmental perspectives in this area.