SEARCH · PubMed Health

Results for “data quality”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 757 records · Page 42Linked to original sources

PubMed2007

An analysis of the exclusion criteria used in observational pharmacoepidemiological studies.

PURPOSE: The application of exclusion criteria in pharmacoepidemiological studies could have a major impact on the findings but there appears to have been no previous research to examine the types of exclusion criteria applied. METHODS: We searched the literature and identified 10 senior pharmacoepidemiologists who had published five or more relevant papers between 1999 and 2004. All their published drug safety studies during this period were reviewed. A classification system was developed to categorise the exclusion criteria, with 5 categories and 11 sub-categories. The categories were: (1) data quality and validation, (2) disease-related, (3) exposure-related, (4) patient characteristics and (5) miscellaneous reasons. Within each sub-category, only the first exclusion criterion identified for that study was counted. RESULTS: We identified 200 studies, from which a total of 752 exclusion criteria sub-categories had been applied (mean 3.8 per study; between-author range of means 2.8-5.1). At the category level, exclusion criteria relating to data quality and validation were the most commonly applied (87% of publications), followed by patient characteristics (75%), disease-related (69%), exposure-related (38%) and miscellaneous (3%). The main categories for which research practice appeared to differ were those relating to diseases and exposures. The application of sub-category 'risk factors and alternative causes' varied between authors from 0% to 81% of studies, and for the sub-category 'medication of interest' it varied from 5% to 93%. CONCLUSIONS: There are important differences between investigators in the application of exclusion criteria in pharmacoepidemiological studies. It is likely that a substantial part of the observed variation reflects different research practices of investigators.

Adverse Drug Reaction Reporting Systems↗

PubMed2000

The SWAL-QOL outcomes tool for oropharyngeal dysphagia in adults: II. Item reduction and preliminary scaling.

The SWAL-QOL outcomes tool was constructed for use in clinical research for patients with oropharyngeal dysphagia. The SWAL-QOL was constructed a priori to enable preliminary psychometric analyses of items and scales before its final validation. This article describes data analysis from a pretest of the SWAL-QOL. We evaluated the different domains of the SWAL-QOL for respondent burden, data quality, item variability, item convergent validity, internal consistency reliability as measured by Cronbach's alpha, and range and skewness of scale scores upon aggregation and floor and ceiling effects. The item reduction techniques outlined reduced the SWAL-QOL from 185 to 93 items. The pretest of the SWAL-QOL afforded us the opportunity to select items for the ongoing validation study which optimally met our a priori psychometric criteria of high data quality, normal item distributions, and robust evidence of item convergent validity.

Aged↗

PubMed2005

Salines, osmoles and albumin.

In 2004, two large randomised multi-centre Australian clinical trials provided new information concerning optimal resuscitation for patients with traumatic brain injury (TBI). One examined hypertonic saline (HTS) and the other, albumin versus saline.( )For the first time in a randomised trial, hypertonic saline was tested for pre-hospital resuscitation of hypotensive patients with traumatic brain injury, and for the first time a resuscitation fluid trial measured long term neurological function as the primary outcome. Despite many potential advantages which may have much greater relevance in the hospital setting, in the paramedic based VICn trauma system, HTS did not improve neurological outcome compared to conventional pre-hospital fluid protocols. Nevertheless, HTS resuscitation was confirmed to be safe in TBI patients and may find application in future pre-hospital military settings where fluid weight is of primary importance. The very large randomised SAFE trial found that there was no difference in 28 day survival between albumin and saline resuscitation for intensive care patients, and by providing very high quality data, this study has largely settled a generation old controversy. Intriguingly however, the SAFE study also reported that within a subgroup of 492 patients with TBI, 28 day survival was superior in patients receiving saline. This subgroup result was not considered definitive, but a post hoc examination of the TBI patients currently in progress by the SAFE investigators, is expected to provide further guidance for clinicians. In the meanwhile, and until more high quality data is available, many clinicians are likely to prefer crystalloid resuscitation for trauma patients, and especially for trauma patients with brain injury.

Journal Article↗

PubMed1992

Need for and influence of feedback from the Finnish birth register to data providers.

Data for the Finnish medical birth register (established 1987) are collected by local hospital personnel as a part of their routine work. The purpose of this study was to study the need of personnel for feedback and the impact of feedback on later data quality. Furthermore, we studied whether such feedback tends to modify extreme cesarean section rates. Data on attitudes towards the birth register and on the need for feedback of data providers were collected through interviews and observations. In March 1988, an information package describing births, birth procedures and infant outcomes in each hospital compared with other hospitals was sent to a random stratified sample of 26 hospitals out of a total of 53. Opinions of the package were obtained by questionnaire from 104 physicians and nurses (82% response rate). Most hospital personnel, especially physicians, had negative attitudes towards the birth register. Comparison of the hospitals which had received feedback with other hospitals in terms of quality of data furnished in 1987 and 1988 suggested that feedback may improve the technical quality of data. There was no evidence, however, that feedback caused hospitals to change their practices in regard to cesarean sections.

Attitude of Health Personnel↗

PubMed1996

Monitoring of cerebral oxygenation in patients with severe head injuries: brain tissue PO2 versus jugular vein oxygen saturation.

Monitoring of cerebral oxygenation is considered to be of great importance in minimizing secondary hypoxic and ischemic brain damage following severe head injury. Although the threshold for cerebral hypoxia in jugular bulb oximetry (measurement of O2 saturation in the jugular vein (SjvO2)) is generally accepted to be 50% oxygen saturation, a comparable value in brain tissue PO2 (PtiO2) monitoring, a new method for direct assessment of PO2 in the cerebral white matter, has not yet been established. Hence, the purpose of this study was to compare brain PtiO2 with SjvO2 in severely head injured patients during phases of reduced cerebral perfusion pressure (CPP) to define a threshold in brain PtiO2 monitoring. In addition, the safety and data quality of both SjvO2 and brain PtiO2 monitoring were studied. In 15 patients with severe head injuries, SjvO2 and brain PtiO2 were monitored simultaneously. For brain PtiO2 monitoring a polarographic microcatheter was inserted in the frontal cerebral white matter, whereas for SjvO2 measurements were obtained by using a fiberoptic catheter placed in the jugular bulb. Intracranial pressure was monitored by means of an intraparenchymal catheter. Mean arterial blood pressure, CPP, end-tidal CO2, and arterial oxygen saturation (pulse oximetry) were continuously recorded. All data were simultaneously stored and analyzed using a multimodal computer system. For specific analysis, phases of marked deterioration in systemic blood pressure and consecutive reductions in CPP were investigated. There were no complications that could be attributed to the PtiO2 catheters, that is, no intracranial bleeding or infection. The "time of good data quality" was 95% in brain PtiO2 compared to 43% in SjvO2; PtiO2 monitoring could be performed twice as long as SjvO2 monitoring. During marked decreases in CPP, SjvO2 and brain PtiO2 correlated closely. A significant second-order regression curve of SjvO2 versus brain PtiO2 (p < 0.01) was plotted. At a threshold of 50% in SjvO2, brain PtiO2 was found to be within the range of 3 to 12 mm Hg, with a regression curve "best fit" value of 8.5 mm Hg. There was a close correlation between CPP and oxygenation parameters (PtiO2 and SjvO2) when CPP fell below a breakpoint of 60 mm Hg, suggesting intact cerebral autoregulation in most patients. This study demonstrates that monitoring brain PtiO2 is a safe, reliable, and sensitive diagnostic method to follow cerebral oxygenation. In comparison to SjvO2, PtiO2 is more suitable for long-term monitoring. It can be used to minimize episodes of secondary cerebral maloxygenation after severe head injury and may, hopefully, improve the outcome in severely head injured patients.

Adolescent↗

PubMed2002

Reusable sensors in intrapartum foetal reflection pulse oximetry.

OBJECTIVES: To evaluate the accuracy of intrapartum foetal pulse oximetry (SO(2POX)) using reusable sensors and the effect of a sensor performance test on data quality. Furthermore, to assess the sensor-related costs by using reusable sensors and sensor performance test. METHODS: 36 reusable sensors were used for SO(2POX) during labour of 289 term foetuses. A sensor performance test device assessing the emitter and receiver capability and the firmness of attachment of the sensors had been developed and used in the last 134 measurements before each resterilisation. Oxygen saturation (SaO(2)) at birth was measured spectrophotometrically after cord blood sampling. The accuracy of SO(2POX) was evaluated by analysing its relationship to SaO(2). The valid SO(2POX) data, as confirmed by subsequent sensor test in the second group, was considered comparable with those with single sensor use. Sensor-related average cost (sensors, test device and sterilisation) of such measurements was compared with that of single sensor use. RESULTS: Eight sensors failed performance test despite valid pulse oximetry signal output during their last measurements. There were significant overall linear correlations between SO(2POX) and SaO(2) (r=0.45, P<0.0001). Separate analyses of regression in the group without sensor performance testing showed an r(2) of 0.41, whereas in the group with subsequent sensor performance testing, the r(2) was 0.52 (P<0.05). By reusing the sensors, the sensor-related cost per valid measurement was $18.9 and 71% lower compared to single use of sensors ($65). CONCLUSIONS: Pulse oximetry may reflect fetal oxygen saturation. Data quality may be compromised by insufficient sensor performance, even though the reflection signal quality is acceptable. If sensor performance is tested before each measurement, reusable sensors may reduce the costs of fetal pulse oximetry.

Equipment Reuse↗

PubMed2000

Conservative interventions for treating distal radial fractures in adults.

BACKGROUND: Fracture of the distal radius is a common clinical problem particularly in elderly white women with osteoporosis, in whom the lifetime risk of this injury has been estimated as 15 per cent. OBJECTIVES: To determine the most appropriate conservative treatment for fractures (such as Colles') of the distal radius in adults. SEARCH STRATEGY: We searched the Cochrane Musculoskeletal Injuries Group trials register, Medline and reference lists of trials. Date of the most recent search: April 1998. SELECTION CRITERIA: Randomised or quasi-randomised clinical trials involving skeletally mature patients with a fracture of the distal radius, which compared commonly applied conservative interventions for fracture fixation. These included the application of an external support (plaster cast or brace) and fracture manipulation. DATA COLLECTION AND ANALYSIS: All trials, judged as fitting the selection criteria by both reviewers, were independently assessed by both reviewers for methodological quality. Data were extracted for anatomical, functional and clinical, including complications, outcomes. The trials were grouped into categories relating to cast position, extent and duration of immobilisation, use of braces, cast material and fracture manipulation. Although quantitative data from some trials are presented, the lack of good quality trials and trial heterogeneity inhibited pooling of results. MAIN RESULTS: Over 50 randomised trials comparing treatment options for distal radial fractures were identified. Twenty nine trials of these, involving a total of 3199 mainly female and older patients, met the inclusion criteria for this review. Comprehensive details of the individual trials are provided in tabular form, and their results, grouped as indicated above, have been presented in text and analyses tables. The poor quality and heterogeneity in terms of patient characteristics, interventions compared and outcome measurement, of the included trials meant that no meta-analyses were undertaken. REVIEWER'S CONCLUSIONS: There is insufficient evidence from randomised trials to determine which methods of conservative treatment are the most appropriate for the more common types of distal radial fractures in adults. Therefore, at present, practitioners applying conservative management should use an accepted technique with which they are familiar, and which is cost-effective from the perspective of their provider unit. Whilst tempting, any call for further research to clarify the most appropriate conservative treatment for this common fracture must be resisted for now. Instead it is important to reflect on the issues raised in this review, to ascertain patient preferences, prioritise researchable questions and undertake a systematic programme of research after agreeing a core data set for classification of fractures and outcomes.

Adult↗

PubMed2003

Novel bacterial ratio for predicting faecal age.

This study presents an extension of ongoing research into the utility of the ratio of colonies isolated on membrane filters during the total coliform test using m-Endo broth media. Investigations into the relative shifts in concentrations of indicator bacterial populations over time, in laboratory-based survival studies conducted with filtered river water, were undertaken. Also, analysis of Kentucky River water quality data collected from the inlet of a local water treatment plant was carried out. Survival studies found that the ratio between the raw concentrations of atypical colonies (AC) and total coliform colonies (TC) was directly related to the amount of time coliform spiked river water had been held in open jars in the laboratory. The AC/TC ratio in the jars would rise from < 1 at the time of coliform spiking to > 200 within 4d. The rise in AC/TC ratio with time in river water was confirmed in the analysis of two years of Kentucky River water quality data where the average AC/TC ratio during months with high river flow (rain) was 3.37 and rose to an average of 27.58 during months with low flow. The average AC/TC ratio during high flow months compared to that of raw human sewage (3.9) and the ratio increased to values associated with animal impacted urban runoff (18.9) during low flow months.

Enterobacteriaceae↗

PubMedDate not supplied

Suggestions for the presentation of quality of life data from clinical trials.

Quality of life (QOL) data is complex since it is both multidimensional and longitudinal. This complexity is compounded with its unbalanced nature through missing observations as a consequence of patient non-compliance with assessment schedules, and, for example, in cancer clinical trials data absence due to patient attrition often through death. QOL data poses difficulties for presentation and analysis and hence interpretation. This paper illustrates, using data from a randomized trial of the United Kingdom Medical Research Council Lung Cancer Working Party, a step-by-step approach to presentation of QOL data. This begins with a description of compliance and its relationship with patient attrition caused by death, to a final summary profile to indicate change over time. We recognize that no single summary statistic is likely to be able to encapsulate all the subtleties of QOL data. We stress the importance of examining data graphically before performing detailed analysis and also to facilitate interpretation in the final clinical report. Although a description of analytical methods is not the purpose of this paper, we draw attention to the need for imputing missing values and to the (multi-level) modelling approach to summarizing the data, both essential adjuncts to the less formal methods described here.

Authorship↗

PubMed1987

A comparison of responses to mailed questionnaires and telephone interviews in a mixed mode health survey.

This paper compares respondents to mailed questionnaires with those nonrespondents subsequently interviewed by telephone in a survey of Massachusetts women aged 45-55 years conducted in 1981-1982. This mixed mode approach produced 8,050 responses, giving a response rate of 77%. This rate is similar to rates obtained in many surveys that employed in-person interviews, which are still widely used in health surveys but are increasingly expensive. Telephone respondents differed socioeconomically from mail respondents, suggesting that telephone follow-up of nonrespondents may have reduced nonresponse bias in this survey. Thus, a mixed mode approach may be superior to a mail-only approach with respect to this aspect of data quality. Women responding by mail were more likely to hold professional jobs, to have relatively high household incomes, and to have more years of education. Controlling for these socioeconomic differences did not, however, remove differences in reported health outcomes between mail and telephone respondents. These differences may be explained by less complete recall in the telephone interviews or they may arise from actual differences in health profiles between early (i.e., mail) and late (i.e., telephone) respondents. Although a mixed mode approach may reduce nonresponse bias, more research is required concerning the reasons for response differences between modes and to eliminate any differences caused by problems in data quality.

Female↗

PubMed2005

The combination of liquid chromatography/tandem mass spectrometry and chip-based infusion for improved screening and characterization of drug metabolites.

An approach has been developed for drug metabolism studies of non-radiolabeled compounds using on-line liquid chromatography/tandem mass spectrometry (LC/MS/MS) combined with chip-based infusion following fraction collection. The potential of this approach, which improves the data quality compared with only LC/MS analysis, has been investigated for the analysis of in vitro metabolites of tolcapone and talinolol, two compounds with well-characterized metabolism. The information-dependent LC/MS/MS analysis enables the characterization of the major metabolites while the chip-based infusion is used to obtain good product ion spectra for lower level metabolites, to generate complementary MS information on potential metabolites detected in the LC/MS trace, or to screen for unexpected metabolites. Fractions from the chromatographic analysis are collected in 20 second steps, into a 96-well plate. The fractions of interest can be re-analyzed with chip-based infusion on a variety of mass spectrometers including triple quadrupole linear ion trap (QqLIT or Q TRAP) and QqTOF systems. Acquiring data for several minutes using multi-channel acquisition (MCA), or signal averaging while infusing the fractions at approximately 200 nL/min, permits about a 50 times gain in sensitivity (signal-to-noise) in MS/MS mode. A 5-10 microL sample fraction can be infused for more than 30 min allowing the time to perform various MS experiments such as MS(n), precursor ion or neutral loss scans and accurate mass measurement, all in either positive or negative mode. Through fraction collection and infusion, a significant gain in data quality is obtained along with a time-saving benefit, because the original sample needs neither to be re-analyzed by re-injection nor to be pre-concentrated. Therefore, a novel hydroxylated talinolol metabolite could be characterized with only one injection.

Animals↗

PubMed2003

Multiplexing nuclear receptors for agonist identification in a cell-based reporter gene high-throughput screen.

High-throughput screening (HTS) has become an essential part of the drug discovery process. Due to the rising requirements for both data quality and quantity, along with increased screening cost and the demand to shorten the time for lead identification, increasing throughput and cost-effectiveness has become a necessity in the hit identification process. The authors present a multiplexed HTS for 2 nuclear receptors, the farnesoid X-activated receptor and the peroxisome proliferator-activated receptor delta in a viable cell-based reporter gene assay. The 2 nuclear receptors were individually transfected into human hepatoma cells, and the transient transfected cell lines were pooled for the multiplexed screen. Hits identified by the multiplexed screen are similar to those identified by the individual receptor screens. Furthermore, the multiplexed screen provides selectivity information if ligands selective for one and not the other receptor are one of the hit criteria. The data demonstrate that multiplexing nuclear receptors can be a simple, efficient, cost-effective, and reliable alternative to traditional HTS of individual targets without compromising data quality.

Cell Line, Tumor↗

PubMed1994

Objective data and quality assurance programs. Current and future trends.

As CCM has grown, the diversity of ICU patients, as well as that of ICU organization and structure, has grown. This growth has led to numerous questions regarding health care delivery in the ICU. These questions contributed to the development of systems that objectively evaluate the quality of health care delivery in ICUs. Severity of disease scoring systems have been developed and allow for a valid analysis of ICU performance at several levels. These systems should help intensivists determine how health care delivery can be optimized in ICUs. Despite the controversy that surrounds severity of disease scoring and prognostic systems, it is not unreasonable to suggest that, because of the feedback these systems would provide, health care delivery in the ICU would be improved through more extensive use of them at the present time. The information acquired through the use of objective scoring systems ultimately must be used to improve the efficiency of ICUs. The structure and organization of ICUs in the United States, as well as the training of those who treat ICU patients, are excessively diverse, and a more standardized approach to health care delivery in the ICU ultimately will be required. Present information suggests that decentralized ICUs with part-time ICU physicians result in poorer outcomes. The APACHE III study intends to explore these relationships in more detail. Certainly, more studies looking at these issues are needed, but we are at least beginning to answer the questions that resulted from the rapid growth of critical care in the 1980s. The SCCM data suggest two possible alternatives, not necessarily exclusive of each other: (1) A large percentage of ICUs may be obligated to undergo structural changes in the near future. (2) Regionalization of critical care, already present, may continue. Certain rural areas may find it more expedient to send the most critically ill patients to tertiary centers in nearby cities, as opposed to a wholesale upgrading of the delivery of care in their own ICUs. Ultimately, all hospitals will be obligated to provide patients access to the highest quality of critical care.

Certification↗

PubMedDate not supplied

Improving the quality of data used in environmental economics research.

Valuing environmental goods through contingent valuation and through other survey techniques is difficult because of the lack of traditional markets for those goods. EPA provides leadership in overcoming these challenges, using focus groups, workshops, handbooks, and teams of distinguished economists to obtain more reliable economic data from the surveys. Agency economists have provided leadership in employing production economics tools to identify win-win approaches to solve pervasive and neglected environmental problems. In addition, we exploit the speed of sophisticated computers, building field runoff models into economic models and aggregating results from tens of thousands of field sites. These very flexible models provide much more credible, site-specific analyses than previous models, and they offer flexible and efficient remedies to support Total Maximum Daily Loads and other new programs.

Agriculture↗

PubMed2004

Mixed alkanethiol self-assembled monolayers as substrates for microarraying applications.

In current microarraying experiments, data quality is in large part determined by the quality of the spots that compose the microarray. Since many microarrays are made with contact printing techniques, microarray spot quality is fundamentally linked to the surface characteristics of the microarray substrate. In this work, surface coatings, consisting of self-assembled monolayers (SAMs) of mixed alkanethiol molecules, were used to control the surface properties of the microarray substrate. X-ray photoelectron spectroscopy and equilibrium contact angle measurements were performed in order to confirm the chemical content and wettability of these surface coatings. To test their performance in microarraying applications, sample microarrays were printed on these mixed alkanethiol films and then characterized with a noncontact visual metrology system and a fluorescence scanner. This work demonstrates that utilizing mixed alkanethiol SAMs as a surface coating provides spatially homogeneous surface characteristics that are reproducible across multiple microarray substrates as well as within a substrate. In addition, this paper demonstrates that these films are stable and robust as they can maintain their surface characteristics over time. Overall, it is demonstrated that SAMs of mixed alkanethiols serve as a useful surface coating, which enhances spot and therefore data quality in microarraying applications.

Alkanes↗

PubMed2006

[Implementation of the multi-center basis documentation CL-BaDo for conciliar and liaison services: generating data for internal quality management and cost calculation].

In the German DRG system the funding of CL services is not ensured. The documentation of psychiatric comorbidity and CL care delivery is a pre-condition to the development of funding models for CL-services. A task force of several German psychosomatic associations (German College of Psychosomatic Medicine, German Society of Psychosomatic Medicine and Psychotherapy, General Medical Society for Psychotherapy) developed a new documentation form for CL-services (CL-BaDo). The pilot study explored the multicenter implementation of CL-BaDo and the use of the documentation form for quality management and cost calculation. Over a period of at least three months, participating CL-services documented all CL cases consecutively with the CL-BaDo. One site applied full electronic data processing. 2116 CL cases from eight psychosomatic CL-services were analysed. The CL-BaDo is a time-efficient, feasible and acceptable documentation form for CL-service delivery. The full electronic data processing enables networking with a hospital information system to produce higher data quality. The data of CL-BaDo can be used locally for quality management, development of management strategies and communication with consultants, as well as nationwide for health policy questions and research.

Austria↗

PubMed2002

Metabolic control and quality of data in medical records for subjects with type 2 diabetes in Swedish primary care: improvement between 1995 and 2001.

OBJECTIVE: To compare the registering of diabetic complications and metabolic control in type 2 diabetic subjects in primary health care in 1995 and 2001. DESIGN: Cross-sectional surveys of medical records. SETTING: Two primary health care centres in Stockholm County. SUBJECTS: Diabetic patients aged 35-64 years; in 1995, 59 men and 42 women and in 2001, 80 men and 80 women. MAIN OUTCOME MEASURES: Rate of notified examinations in medical records, diabetic complications and metabolic control. RESULTS: Increased rates between 1995 and 2001 were found regarding examinations for neuropathy (46% vs 79%), albuminuria (74% vs 91%), micro-albuminuria (11% vs 59%), BMI (33% vs 66%), HbA1c (88% vs 99%), cholesterol (69% vs 92%) and LDL cholesterol (42% vs 89%) (p < 0.001). Increased rates of perceived treatment goals between 1995 and 2001 were found regarding HbA1c (< 6.5%) (28% vs 57%) and cholesterol (< 5.0 mmol/l) (23% vs 36%) (p < 0.001), and also regarding LDL cholesterol (30% vs 36%) (p < 0.01). In men, the number of subjects with any macrovascular complication decreased between 1995 and 2001 (34% vs 17%) (p < 0.01), and in women the number of subjects with coronary heart disease decreased (26% vs 6%) (p < 0.01). CONCLUSION: Documented examination rates and metabolic control improved between 1995 and 2001.

Adult↗

PubMed2000

Vitamin K antagonists or low-molecular-weight heparin for the long term treatment of symptomatic venous thromboembolism.

BACKGROUND: Patients who have had an episode of symptomatic venous thromboembolism are usually treated for at least five days with intravenous unfractionated heparin or subcutaneous low-molecular-weight heparin. Thereafter, they received a three month course of a vitamin K antagonist, with a dose adjusted to achieve an International Normalized Ratio between 2.0 and 3.0. Some patients have contraindications to vitamin K antagonists. In addition, treatment with vitamin K antagonists has the disadvantage of regular laboratory measurements. OBJECTIVES: The objective of this review was to evaluate the efficacy and safety of long-term treatment of symptomatic venous thromboembolism with low-molecular-weight heparins compared with vitamin K antagonists. SEARCH STRATEGY: Computerized searches of MEDLINE, EMBASE and Current Contents were made and relevant journals were hand-searched using the search strategy described by the Cochrane Peripheral Vascular Disease Group. In addition, randomized clinical trials were located through personal communication with colleagues. Where necessary, the reviewers contacted pharmaceutical companies for further information. SELECTION CRITERIA: Two reviewers evaluated studies independently for methodological quality. DATA COLLECTION AND ANALYSIS: Two reviewers reviewed and extracted data independently using a standard form. Primary analysis concerned all patients in the studies during the period of randomized treatment. Additional separate analyses were performed for category I and category II studies; studies that used similar initial treatments in both study arms and those that used different treatment regimes during the initial treatment; and the total period of follow-up in the different studies. MAIN RESULTS: Five studies were identified that fulfilled our predefined criteria (three category I and two category II studies). When all five studies were combined, a statistically non-significant reduction of the risk of recurrent symptomatic venous thromboembolism in favor of low-molecular-weight heparin treatment (OR 0.72; 95% CI [0.42, 1.23]) was found. In category I studies, analysis of the pooled data showed a statistically non-significant reduction of the risk of recurrent symptomatic venous thromboembolism in favor of low-molecular-weight heparin treatment (OR 0.75; 95% CI [0.40, 1.39]). This OR was mainly due to one possibly confounded study, and after omitting this study from the analysis a statistically non-significant reduction of the risk of recurrent symptomatic venous thromboembolism in favor of vitamin K antagonist treatment remained (OR 1.95; 95% CI [0.74, 5.19]). No differences in the risk of bleeding (OR 0.63; 95% CI [0.21, 1.88]) and mortality (OR 1.13; 95% CI [0.47, 2.69] were observed. REVIEWER'S CONCLUSIONS: Low-molecular-weight heparins are possibly as effective and safe as vitamin K antagonists in the prevention of recurrent symptomatic venous thromboembolism after an episode of symptomatic deep venous thrombosis, but have the disadvantage of much higher medicinal costs. Treatment with low-molecular-weight heparin is possibly a safe alternative in some patients; for example patients who live in geographically inaccessible places; patients who are reluctant to go to the thrombosis service on a regular basis; and patients with contraindications to vitamin K antagonists (e.g. pregnant women). Therefore, in the absence of definitive evidence on the safety and efficacy of low-molecular-weight heparins compared with vitamin K antagonists, we believe that treatment with vitamin K antagonists is still the treatment of choice in the prevention of recurrent symptomatic venous thromboembolism after an episode of deep venous thrombosis, in the majority of patients.

Anticoagulants↗