Paper Charts Research Articles

The implementation of a streamlined process for biomarker testing in medical oncology.

179 Background: Biomarker testing is increasingly becoming an essential part of standard care. At Sunnybrook Health Sciences Centre (SHSC) in Toronto, Canada, our Department of Molecular Services handles some internal biomarker tests, while some tests are referred to citywide labs. The Odette Cancer Centre (OCC) of SHSC is one of the largest cancer centres in Canada, serving over 12,000 new patients per year, of whom many rely on biomarker tests for personalized treatment. We aimed to describe the work systems at the OCC for biomarker testing and reporting, and to initiate system improvements. Methods: This quality improvement initiative occurred in three phases: qualitative descriptive analysis of the current process of biomarker testing, exploration of future state process using LEAN, and implementation of a streamlined process. In phase one, ten medical oncologists, two administrative assistants, and one pathologist were interviewed. A multidisciplinary team was then assembled to investigate and initiate improvements. Results: Tracking results from external labs was managed by individual physicians and hard copy results were submitted to medical records for filing in patient paper charts, compared to internal tests which are posted on the electronic record, making outside tests harder to retrieve later. The current process involved more than 150 different tests with only 44% of results appearing in the hospital electronic record. In June 2016, a standardized process was implemented where a designated laboratory assistant managed requisition forms, sent corresponding specimens to qualified labs, ensured the receipt of results through various electronic tracking tools and validated the subsequent upload to the electronic medical system. Over a four-month implementation period, there were 364 cases/patients with 467 tests requested; 100% of these test results are stored in the electronic record. Conclusions: The lack of standardization of biomarker testing and reporting can have negative implications on quality of care and patient safety. Therefore, streamlining this process and incorporating electronic tracking tools can improve the accessibility of test results to improve the quality of oncology care.

Electronic medical records in humanitarian emergencies - the development of an Ebola clinical information and patient management system.

By November 2015, the West Africa Ebola epidemic had caused 28598 infections and 11299 deaths in the three countries most affected. The outbreak required rapid innovation and adaptation. Médecins sans Frontières (MSF) scaled up its usual 20-30 bed Ebola management centres (EMCs) to 100-300 beds with over 300 workers in some settings. This brought challenges in patient and clinical data management resulting from the difficulties of working safely with high numbers of Ebola patients. We describe a project MSF established with software developers and the Google Social Impact Team to develop context-adapted tools to address the challenges of recording Ebola clinical information. We share the outcomes and key lessons learned in innovating rapidly under pressure in difficult environmental conditions. Information on adoption, maintenance, and data quality was gathered through review of project documentation, discussions with field staff and key project stakeholders, and analysis of tablet data. In March 2015, a full prototype was deployed in Magburaka EMC, Sierra Leone. Inpatient data were captured on 204 clinical interactions with 34 patients from 5 March until 10 April 2015. Data continued to also be recorded on paper charts, creating theoretically identical record "pairs" on paper and tablet. 83 record pairs for 33 patients with 22 data items (temperature and symptoms) per pair were analysed. The overall Kappa coefficient for agreement between sources was 0.62, but reduced to 0.59 when rare bleeding symptoms were excluded, indicating moderate to good agreement. The time taken to deliver the product was more than that anticipated by MSF (7 months versus 6 weeks). Deployment of the tablet coincided with a dramatic drop in patient numbers and thus had little impact on patient care. We have identified lessons specific to humanitarian-technology collaborative projects and propose a framework for emergency humanitarian innovation. Time and effort is required to bridge differences in organisational culture between the technology and humanitarian worlds. This investment is essential for establishing a shared vision on deliverables, urgency, and ownership of product.

National Rates and Patterns of Depression Screening in Primary Care: Results From 2012 and 2013.

Despite high prevalence rates of depression in primary care, depressive symptoms are often undetected by physicians. Screening for depression is now recommended as a part of routine primary care; however, recent estimates of rates and patterns of depression screening are lacking in the literature. This study examined national rates and patterns of depression screening among visits to office-based primary care physicians. A secondary analysis of data from the 2012 and 2013 National Ambulatory Medical Care Survey was conducted. The sample consisted of 33,653 physician-patient encounters. The overall rate of depression screening was 4.2%. African Americans were half as likely to be screened compared with whites, and elderly patients were half as likely to be screened compared with middle-aged patients. Patients with a chronic condition were more likely than patients without a chronic condition to receive depression screening, and the likelihood of being screened increased with each additional chronic condition. Providers who had fully adopted electronic health records (EHRs) were more likely to screen for depression compared with providers who used paper charts. Screening rates were not associated with providers' intentions to participate in the federal program that provides financial incentives for the meaningful use of certified EHRs. Overall rates of depression screening were low. Current screening practices may exacerbate existing disparities in depression care. EHR systems may be an effective tool to improve screening rates.

Safety Outcomes of Apixaban Compared With Warfarin in Patients With End-Stage Renal Disease.

Current guidelines make no specific recommendations on the selection of direct oral anticoagulants for the prevention and treatment of venous thromboembolism in patients with end-stage renal disease (ESRD) receiving hemodialysis. Based on these guidelines, warfarin remains the anticoagulant of choice in these patients. To compare bleeding rates in patients receiving apixaban or warfarin with ESRD undergoing chronic hemodialysis. This was a single-center, retrospective, institutional review board-approved cohort analysis. Patients with ESRD undergoing chronic hemodialysis and receiving anticoagulation therapy with either apixaban or warfarin were included in this study. All data were collected from paper charts and electronic medical records and included documentation of bleeding events and related interventions. The primary outcome of this study was clinically relevant major bleeding events. Secondary outcomes included clinically relevant nonmajor bleeding events and minor bleeding events. A total of 160 patients were included in this study (warfarin group, n = 120; apixaban group, n = 40). There were 7 major bleeding events in the warfarin group compared with zero in the apixaban group ( P = 0.34). There were similar rates of clinically relevant nonmajor bleeding events (12.5% vs 5.8%, P = 0.17) and minor bleeding (2.5% vs 2.5%, P = 0.74) events in patients receiving apixaban and warfarin. There were no observed differences in bleeding rates in patients receiving apixaban compared with those receiving warfarin. Apixaban may be a cautious consideration in hemodialysis patients until there is further insight into the effect of subsequent, multiple doses on drug accumulation and clinical outcomes.

Suicide Risk Assessment by Residents: Deficiencies of Documentation.

The authors sought to assess the documentation of suicide risk assessments performed by psychiatry residents in a psychiatric emergency service (PES) and to identify differences in documentation between previously used paper charts and a new electronic medical record (EMR) system based on the Columbia Suicide Severity Rating Scale (C-SSRS)-risk assessment version. This study is a retrospective chart review of psychiatric evaluations performed by psychiatry residents during a 1-year period in the PES of a large, urban, academic medical center. The sample was selected by a systematic random sampling technique from a total of 3931 evaluations performed on adult patients during the study period. The suicide risk assessments were evaluated using data regarding demographics, process indicators identified from the C-SSRS, and diagnoses. A total of 300 charts were reviewed. Only 91% of the evaluations contained documentation of suicidal ideations (either admitted or denied); 5 other variables were documented in more than 50% of the evaluations: treatment status (95.3%), presence/absence of previous suicide attempts (84.6%), recent event-denies (56%), history of recent negative events (55%), and suicidal behavior-denies (53%). Additionally, 2 risk factors and 3 protective factors were documented in over 25% of the evaluations. Documentation was deficient in multiple areas, with even the presence/absence of suicidal ideations not being documented in all evaluations. Use of an EMR with built-in "clickable" options selectively improved documentation especially regarding risk and protective factors adapted from the C-SSRS. Emphasis on documentation of assessments is paramount while training residents in suicide risk assessment.

A Study on FIDO Authentication System for Reinforcing the Security of Electronic Medical Records

The target of compulsory certification in Information Security Management System has extended to medical institutions. This caused us to recognize the importance of information security in modern hospital information system that has changed from the medical record management that was recorded and managed largely in paper chart in the past to the Electronic Medical Record that medical personnel enter patient information into a computer directly for building a database. As medical institutions manage sensitive information like personal information basically, personal medical data infringement accident, if occurred can become a big social issue. Currently, the medical information in medical institutions are stored in electronic medical records and to access, user authentication is required by means of accredited certificate as security measure. Accredited certification has technical problems such as certificate storage method and security level of password and managerial problems such as certificate copy/leak/share. In this respect, this study proposes and presents how to build the FIDO-based authentication system that applies UAF or U2F authentication mechanism depending on the authority and work scope of medical personnel and medical support assistant like staffs, officers, licensed practical nurse and so on, within large medical institutions that use medical information system. The aim is to solve problems in accredited certificate authentication method in the existing medical institutions with the FIDO-based authentication system proposed in this study.

The Truth About Authenticity

The Truth About Authenticity

FRAIL-NH Simplifies Frailty Screening

FRAIL-NH Simplifies Frailty Screening

An Ethical Approach to Email and Text in Patient Care

An Ethical Approach to Email and Text in Patient Care

The Utility of Mobile Application Technology to Analyze Efficiency And Improve Operational Performance in the Endoscopy Unit

Introduction: The principles of operations management and healthcare information technology (IT) may be used together to evaluate and enhance operational performance in endoscopy centers. Previously, our group published the largest publically available endoscopy unit analysis. This “time and motion” assessment was done via data collection using a paper chart flowsheet.(1) In this study, we aim to implement a mobile app software solution and test its ability to gather endoscopy unit data and analyze efficiency in the endoscopy unit. Ultimately, we hope to validate the use of a mobile app to analyze endoscopy unit efficiency in order to perform targeted operational interventions to reduce overhead cost savings. Methods: The study was conducted over a 5-week period at a UCLA outpatient community-based endoscopy center. Endoscopy unit staff members were each given an iPad device to input operational check points involving patient flow through the endoscopy unit in real time. A variety of performance metrics were tracked for efficiency, including the True Completion Time (TCT) (scheduled procedure start time to the time the patient exited the room post-procedure), on-time starts (endoscope entry occurring within10 minutes of the scheduled time of procedure), and mean delay (time from this 10-minute mark to endoscope entry) for all cases studied. Results: We analyzed 168 patients who underwent a total of 249 procedures during the study period [42 EGD, 154 Colonoscopy, 51 EGD+Colonoscopy, 2 Flexible Sigmoidoscopy]. A total of 5 endoscopists were included in the study. The TCT was 47 minutes and room turnover time was 10 minutes across all cases studied. Ninety three percent of procedures had on-time starts, with a mean delay of 2 minutes. The average procedure duration was 39 minutes. Conclusion: The use of mobile app technology is a feasible method to implement real-time tracking and management of operational efficiency in the endoscopy center. This technique of efficiency analysis offers potential advantages compared to paper charting including automation and the potential for real-time data analysis and workflow modifications. This software-based technology may facilitate the implementation of large-scale operational efficiency and process improvement projects aimed at enhancing workflow. (1)Kaushal, et al. “Using efficiency analysis and targeted intervention to improve operational performance and achieve cost savings in the endoscopy center.” Gastrointest Endosc 2014;79:637-45.Table 1: Average Time Benchmarks, all cases studiedFigure 1

Electronic medical records in humanitarian emergencies – the development of an Ebola clinical information and patient management system

By November 2015, the West Africa Ebola epidemic had caused 28598 infections and 11299 deaths in the three countries most affected. The outbreak required rapid innovation and adaptation. Médecins sans Frontières (MSF) scaled up its usual 20-30 bed Ebola management centres (EMCs) to 100-300 beds with over 300 workers in some settings. This brought challenges in patient and clinical data management resulting from the difficulties of working safely with high numbers of Ebola patients. We describe a project MSF established with software developers and the Google Social Impact Team to develop context-adapted tools to address the challenges of recording Ebola clinical information. We share the outcomes and key lessons learned in innovating rapidly under pressure in difficult environmental conditions. Information on adoption, maintenance, and data quality was gathered through review of project documentation, discussions with field staff and key project stakeholders, and analysis of tablet data. In March 2015, a full prototype was deployed in Magburaka EMC, Sierra Leone. Inpatient data were captured on 204 clinical interactions with 34 patients from 5 March until 10 April 2015. Data continued to also be recorded on paper charts, creating theoretically identical record "pairs" on paper and tablet.85 record pairs for 32 patients with 26 data items (temperature and symptoms) per pair were analysed. The average agreement between sources was 85%, ranging from 69% to 95% for individual variables. The time taken to deliver the product was more than that anticipated by MSF (7 months versus 6 weeks). Deployment of the tablet coincided with a dramatic drop in patient numbers and thus had little impact on patient care. We have identified lessons specific to humanitarian-technology collaborative projects and propose a framework for emergency humanitarian innovation. Time and effort is required to bridge differences in organisational culture between the technology and humanitarian worlds. This investment is essential for establishing a shared vision on deliverables, urgency, and ownership of product.

Electronic medical records in humanitarian emergencies – the development of an Ebola clinical information and patient management system

By November 2015, the West Africa Ebola epidemic had caused 28598 infections and 11299 deaths in the three countries most affected. The outbreak required rapid innovation and adaptation. Médecins sans Frontières (MSF) scaled up its usual 20-30 bed Ebola management centres (EMCs) to 100-300 beds with over 300 workers in some settings. This brought challenges in patient and clinical data management resulting from the difficulties of working safely with high numbers of Ebola patients. We describe a project MSF established with software developers and the Google Social Impact Team to develop context-adapted tools to address the challenges of recording Ebola clinical information. We share the outcomes and key lessons learned in innovating rapidly under pressure in difficult environmental conditions. Information on adoption, maintenance, and data quality was gathered through review of project documentation, discussions with field staff and key project stakeholders, and analysis of tablet data. In March 2015, a full prototype was deployed in Magburaka EMC, Sierra Leone. Inpatient data were captured on 204 clinical interactions with 34 patients from 5 March until 10 April 2015. Data continued to also be recorded on paper charts, creating theoretically identical record “pairs” on paper and tablet. 83 record pairs for 33 patients with 22 data items (temperature and symptoms) per pair were analysed. The overall Kappa coefficient for agreement between sources was 0.62, but reduced to 0.59 when rare bleeding symptoms were excluded, indicating moderate to good agreement. The time taken to deliver the product was more than that anticipated by MSF (7 months versus 6 weeks). Deployment of the tablet coincided with a dramatic drop in patient numbers and thus had little impact on patient care. We have identified lessons specific to humanitarian-technology collaborative projects and propose a framework for emergency humanitarian innovation. Time and effort is required to bridge differences in organisational culture between the technology and humanitarian worlds. This investment is essential for establishing a shared vision on deliverables, urgency, and ownership of product.

AN AUDIT OF WHETHER PRESCRIBED DOSES ARE MEASURABLE ON THE GRADUATIONS OF ONE ORAL SYRINGE

AimIn paediatrics drugs are prescribed as mg/kg doses to facilitate accurate dosing. Anecdotally, some drugs are prescribed in such a way that the volume to be given is difficult to...

Opportunities and challenges in conducting secondary analysis of HIV programmes using data from routine health information systems and personal health information.

IntroductionHIV programme data from routine health information systems (RHIS) and personal health information (PHI) provide ample opportunities for secondary data analysis. However, these data pose unique opportunities and challenges for use in health system monitoring, along with process and impact evaluations.MethodsAnalyses focused on retrospective case reviews of four of the HIV-related studies published in this JIAS supplement. We identify specific opportunities and challenges with respect to the secondary analysis of RHIS and PHI data.ResultsChallenges working with both HIV-related RHIS and PHI included missing, inconsistent and implausible data; rapidly changing indicators; systematic differences in the utilization of services; and patient linkages over time and different data sources. Specific challenges among RHIS data included numerous registries and indicators, inconsistent data entry, gaps in data transmission, duplicate registry of information, numerator-denominator incompatibility and infrequent use of data for decision-making. Challenges specific to PHI included the time burden for busy providers, the culture of lax charting, overflowing archives for paper charts and infrequent chart review.ConclusionsMany of the challenges that undermine effective use of RHIS and PHI data for analyses are related to the processes and context of collecting the data, excessive data requirements, lack of knowledge of the purpose of data and the limited use of data among those generating the data. Recommendations include simplifying data sources, analysis and reporting; conducting systematic data quality audits; enhancing the use of data for decision-making; promoting routine chart review linked with simple patient tracking systems; and encouraging open access to RHIS and PHI data for increased use.

Extending the argument from unconceived alternatives: observations, models, predictions, explanations, methods, instruments, experiments, and values

Stanford’s argument against scientific realism focuses on theories, just as many earlier arguments from inconceivability have. However, there are possible arguments against scientific realism involving unconceived (or inconceivable) entities of different types: observations, models, predictions, explanations, methods, instruments, experiments, and values. This paper charts such arguments. In combination, they present the strongest challenge yet to scientific realism.

Acute leukemia in pregnancy: A single institution experience.

7040 Background: Acute leukemia is rarely diagnosed during pregnancy. Management of acute leukemia during pregnancy presents a considerable challenge due to limited available data on optimal management including fetal and maternal outcomes. Methods: After approval by the Mayo IRB, 23 cases of acute leukemia diagnosed during pregnancy between 1962 and 2016 were identified using an institutional computerized database. Information was extracted using electronic medical records and paper chart review. Results: The majority of patients (n=18) had acute myeloid leukemia (AML) including 4 with acute promyelocytic leukemia (APL). Five patients had acute lymphoblastic leukemia (ALL). Ten patients were diagnosed in the first trimester (43.4%), 7 in the second trimester (30.4 %) and 6 in the third trimester (26.2%) of pregnancy. Eighteen patients (78%) achieved a complete remission. Approximately 50% (n=11) of pregnancies ended with elective or spontaneous abortions. Fifty percent (2 /4) of patients diagnosed in first or second trimester who delayed chemotherapy greater than 1 week died during induction therapy. Eleven patients received chemotherapy while pregnant (2 in first, 6 in second and 3 in third trimester) which lead to 4 fetal losses and 7 deliveries. The majority of neonates were of low birth weight with one having mild anemia at birth and no detected congenital malformation. At a median follow up of 55 months (range, 8-330) 7 patients (30%) are alive. Table 1 summarizes the maternal and fetal outcomes. Conclusions: Delay in therapy on the basis of fetal well-being may lead to an adverse maternal outcome. Despite higher risk of preterm low birth-weight delivery, in utero exposure to chemotherapy in the second and third trimesters did not increase the risk of fetal malformation. Overall outcomes based on trimester of pregnancy at the time of acute leukemia diagnosis. N=23 First Trimester 0-13 weeks Second Trimester 14-26 weeks Third Trimester 27-40 weeks N 10 7 6 Median time to chemotherapy (range) 11.5 (2-140) days 4 (1-6) days 1 untreated 6 (1-34) days AML 6 4 4 APL 1 2 1 ALL 3 1 1 Elective Termination 5 0 0 Fetal Outcomes N=5 N=7 N=6 Death 3 3 0 Pre-term Delivery 2 2 4 Term Delivery 0 2 2 Disease Outcomes . . . Achieved CR 8 6 4 Durable CR 3 3 2 Relapsed/Refractory 3 1 2 HSCT 3 1 2

Custodiol versus blood cardioplegia in pediatric cardiac surgery, two-center study

BackgroundCold blood cardioplegia is widely used as a method of myocardial protection in pediatric cardiac surgery. Operative interruption to administer cardioplegia and need for repeated administration and occasional direct coronary ostial cannulation are considered a drawbacks of this technique. Custodiol, a crystalloid solution, has been used in children as an alternative cardioplegia solution with the potential advantage of single dose administration with equivalent myocardial protection. We sought to compare the outcomes of cold blood cardioplegia versus Custodiol cardioplegia in pediatric cardiac surgery. MethodsA retrospective cohort study was performed between November 1st, 2013 and June 30th, 2014. All children who underwent heart surgery at two institutions with the use of cardiopulmonary bypass and cardioplegia were included. Patients were identified from a prospective database and additional data were collected from operative report and electronic and paper charts. Continuous data were represented as median and interquartile range, and tested with Wilcoxon rank-sum test. Categorical data were represented as proportions, and tested with Pearson test. A composite endpoint of all cause death, Low Cardiac Output Syndrome (LCOS), Acute Kidney Injury (AKI), and significant arrhythmia was analyzed with a multiple logistic regression model adjusted for complexity using the Risk Adjustment of Congenital Heart Surgery -1 (RACHS-1) categories. A p-value of less than 0.05 was considered to be significant. ResultsBlood cardioplegia was administered in 88 (57.1%) patients, and Custodiol cardioplegia was administered in 66 (42.9%) patients according to surgeon's preference. In the risk adjusted comparison of the composite outcome of all-cause death, LCOS, AKI and significant arrhythmia, Custodiol cardioplegia was found to be an independent predictor of an adverse outcome, OR 3.17 (95% CI 1.41–7.14, P-value = 0.0054). ConclusionsCustodial cardioplegia is associated with less optimal myocardial protection and higher adverse outcomes compared to cold blood cardioplegia in children undergoing cardiac surgery. A randomized comparison is warranted.

The Impact of the Transition to an Electronic Medical Record on Patient Perceptions in a Pediatric Ophthalmology Practice.

To assess the impact of the transition from traditional paper-based medical records to electronic medical records in a pediatric ophthalmology practice at a tertiary care center. A prospective, cross-sectional survey was completed at three time points: 2 weeks prior to (phase 1), 2 weeks after (phase 2), and 3 months after (phase 3) the electronic medical record transition. The survey consisted of 10 Likert-type scaled questions assessing patient satisfaction and two free response questions estimating the wait time, which was completed by patients or parents/guardians whose child/children (younger than 18 years) had an appointment in the pediatric ophthalmology and strabismus clinic. Satisfaction scores and waiting times were compared within each phase and across phases and between different appointment types. A total of 382 surveys were collected: 158 from phase 1, 68 from phase 2, and 156 from phase 3. Overall, patient satisfaction was high at all three time points. Patients' estimates of waiting time compared to actual waiting time were not significantly different at any phase; however, patients' estimates of time spent with the physician were significantly underestimated in phase 1 (20 vs 25 minutes, P = .04) and were correct or overestimated in phase 3. Patients were satisfied with the service regardless of the use of paper charts or electronic medical records. The electronic medical record system does not seem to improve patients' waiting time, but has a significant impact on the perception of time spent with the physician. [J Pediatr Ophthalmol Strabismus. 2016;53(3):173-178.].

The Role of Hospital Inpatients in Supporting Medication Safety: A Qualitative Study.

BackgroundInpatient medication errors are a significant concern. An approach not yet widely studied is to facilitate greater involvement of inpatients with their medication. At the same time, electronic prescribing is becoming increasingly prevalent in the hospital setting. In this study we aimed to explore hospital inpatients’ involvement with medication safety-related behaviours, facilitators and barriers to this involvement, and the impact of electronic prescribing.MethodsWe conducted ethnographic observations and interviews in two UK hospital organisations, one with established electronic prescribing and one that changed from paper to electronic prescribing during our study. Researchers and lay volunteers observed nurses’ medication administration rounds, pharmacists’ ward rounds, doctor-led ward rounds and drug history taking. We also conducted interviews with healthcare professionals, patients and carers. Interviews were audio-recorded and transcribed. Observation notes and transcripts were coded thematically.ResultsPaper or electronic medication records were shown to patients in only 4 (2%) of 247 cases. However, where they were available during patient-healthcare professional interactions, healthcare professionals often viewed them in order to inform patients about their medicines and answer any questions. Interprofessional discussions about medicines seemed more likely to happen in front of the patient where paper or electronic drug charts were available near the bedside. Patients and carers had more access to paper-based drug charts than electronic equivalents. However, interviews and observations suggest there are potentially more significant factors that affect patient involvement with their inpatient medication. These include patient and healthcare professional beliefs concerning patient involvement, the way in which healthcare professionals operate as a team, and the underlying culture.ConclusionPatients appear to have more access to paper-based records than electronic equivalents. However, to develop interventions to increase patient involvement with medication safety behaviours, a wider range of factors needs to be considered.

Twelve-year survival of 2-surface composite resin and amalgam premolar restorations placed by dental students

Statement of problemComposite resin and amalgam restorations are indicated for the restoration of posterior teeth. With increased esthetic demands, long-term clinical studies are required to evaluate the restorative success and reasons for failure of these materials. PurposeThe purpose of this retrospective study was to determine the survival and reasons for failure of directly placed 2-surface composite resin restorations and directly placed 2-surface amalgam restorations on premolars placed by Canadian dental students. Material and methodsUsing The University of Manitoba’s dental management software and paper charts, all 2-surface composite resin and 2-surface amalgam restorations placed on premolars between January 1, 2002, and May 30, 2014, were included. Short-term failure (within 2 years), long-term failure, and reasons for failure were collected. A Kaplan-Meier survival estimate with an associated P value comparing composite resin to amalgam restoration curves was performed using SPSS statistical software. ResultsOver 12 years, 1695 composite resin and 1125 amalgam 2-surface premolar restorations were placed. Of these restorations, 134 composite resins (7.9%) and 66 amalgams (5.9%) failed. Short-term failures (2 years or less) consisted of 57 composite resin (4%) and 23 amalgam (2.3%) restorations. Long-term failures (greater than 2 years) consisted of 77 composite resin (4.5%) and 43 amalgam (3.8%) restorations. After 12 years of service, the survival probability of composite resin restorations was 86% and that of amalgam restorations 91.5%. The differences in composite resin and amalgam survival curves were also found to be statistically significant (P=.009 for Log-rank test). The main reasons for failure were recurrent caries and fracture of the tooth being restored. ConclusionsWithin the limitations of this study, both composite resin and amalgam restorations had acceptable success rates and similar failure modes. Recurrent caries was still the most common reason for failure.