Computer-based Surveillance System Research Articles

Validation of the WINROP screening algorithm among preterm infants in East Malaysia

Objective: To prospectively validate the WINROP (Weight, Insulin-like growth factor 1, Neonatal, Retinopathy of Prematurity) screening algorithm (www.winrop.com) based on longitudinal measurements of neonatal body weights in predicting the development of severe retinopathy of prematurity (ROP) among preterm infants admitted to the neonatal intensive care unit of a tertiary care center in East Malaysia.
 Methods: All premature infants of less than 32 weeks gestational age (GA) were included in this cohort. Their body weight was measured weekly from birth to 36 weeks postmenstrual age and entered into the computer-based surveillance system: WINROP. Infants were then classified by the system into high- or low-risk alarm group. The retinopathy findings were recorded according to Early Treatment for ROP criteria. However, the screening and management of infants were done according to the recommendations of the Continuous Practice Guidelines, Ministry of Health, Malaysia. The team members involved in screening and those recording the findings were kept blinded from each other.
 Results: A total of 151 infants with median GA at birth of 30 weeks (interquartile range [IQR] Å} 2.1) and mean birth weight of 1,264 g (standard deviation Å} 271) were analyzed. High-risk alarm was signaled in 85 (56.3%) infants and 9 (6.6%) infants developed type 1 ROP. One infant in the low-risk alarm group developed type 1 ROP requiring laser retinal photocoagulation. The median time lag from the high-risk alarm signal to the development of type 1 ROP was 10.4 (IQR Å} 8.4) weeks.
 Conclusion: In this cohort, the WINROP algorithm had a sensitivity of 90%, with negative predictive value of 98.5% (95% confidence interval) for detecting infants with type 1 ROP and was able to predict infants with ROP earlier than their due screening date. This study shows that a modified version of the WINROP algorithm aimed at specific populations may improve the outcome of this technique.

Efficacy of the Screening Algorithm WINROP in a Korean Population of Preterm Infants

To investigate the efficacy of WINROP (https://winrop.com), an algorithm based on serial measurements of neonatal body weight to predict proliferative retinopathy of prematurity (ROP), in a Korean population of preterm infants. The records of preterm infants with gestational age less than 32 weeks who were admitted to the neonatal intensive care unit at Chonnam National University Hospital, Gwangju, South Korea, from October 2006 to November 2010 were reviewed. The body weight of infants was measured weekly and entered into a computer-based surveillance system, WINROP, and the outcome was analyzed. A total of 314 preterm infants participated in the study. The mean gestational age was 29 weeks (range, 25-32 weeks). The mean body weight was 1263 g (range, 590-2260 g). For 166 of 314 infants (52.9%), a high-risk alarm was noted. In the high-risk alarm group, 36 infants developed type 1 ROP, according to the Early Treatment for Retinopathy of Prematurity criteria, and they were treated for ROP. The remaining 148 infants (47.1%) had a low-risk alarm. In the low-risk alarm group, 3 infants with bronchopulmonary dysplasia and intraventricular hemorrhage, a risk factor for ROP, and 1 infant without any risk factors for ROP developed type 1 ROP and were treated. In a Korean population, the WINROP algorithm had a sensitivity of 90% for identifying infants with type 1 ROP. Although some limitations are present, adjustment to the WINROP algorithm for a specific population may improve the efficacy of predicting proliferative ROP and reduce the frequency of retinal examinations.

Surveillance of selected post-caesarean infections based on electronic registries: validation study including post-discharge infections

The importance of surveillance of post-discharge infections has increased as a consequence of shorter hospital stay after surgical procedures. This study examined the ability of a computer-based surveillance system to identify urinary tract infections (UTIs) and postoperative wound infections (PWIs) within 30 days after caesarean section. We assessed the use of data from various electronic registries to identify patients with post-caesarean UTI and PWI classified according to a reference standard. The standard was based on information from medical records and self-reported data (questionnaire) using modified Centers for Disease Control and Prevention definitions. The sensitivity of the computer system in detecting UTI diagnosed during hospital stay, readmission or at visits to hospital outpatient clinics was 80.0%; the specificity was 99.9%. For post-discharge UTIs diagnosed outside the hospital, sensitivity and specificity were 76.3% and 99.9%, respectively. For PWIs diagnosed in hospital and post-discharge outside hospital, sensitivities were 77.1% and 68.9%, and the specificities 99.5% and 98.2%. We conclude that a computer-based surveillance system may identify in-hospital infections and post-discharge infections with a relatively high sensitivity and excellent specificity.

Longitudinal postnatal weight measurements for the prediction of retinopathy of prematurity.

To validate longitudinal postnatal weight gain as a method for predicting severe retinopathy of prematurity (ROP) in a US cohort. Both ROP evaluations and weekly weight measurements from birth to postmenstrual week 36 for 318 infants were entered into a computer-based surveillance system, WINROP. This system signaled an alarm when the rate of weight gain decreased compared with control subjects. Infants were classified into 3 groups: (1) no alarm, (2) low-risk alarm, or (3) high-risk alarm. Maximum ROP for each infant was categorized as (1) no ROP (immature or mature vascularization), (2) mild ROP (stage 1 or 2 ROP in zone II or III, without plus disease), or (3) severe ROP (any prethreshold, any stage 3, or threshold ROP). A high-risk alarm identified infants at risk for developing severe ROP. A high-risk alarm occurred in 81 infants (25.5%) and detected all infants who developed severe ROP a median of 9 weeks before diagnosis. The remaining infants received no alarm or a low-risk alarm. None of these infants developed more than mild ROP. Longitudinal postnatal weight gain may help predict ROP. In a US cohort, the WINROP system had a sensitivity of 100% and identified infants early who developed severe ROP. With further validation, WINROP has the potential to safely reduce the number of ROP examinations.

Implementation of a novel on-ward computer-assisted surveillance system for device-associated infections in an intensive care unit

For the presented study a computer-based surveillance system for detecting nosocomial infections (NI) with direct data input from attending on-ward physicians was implemented. During a 12-month period surveillance of ventilator-associated pneumonia (VAP) and catheter-associated bloodstream infections (BSI) was performed prospectively by on-ward physicians guided by infection control specialists on an 11-bed medical intensive care unit in a German university hospital. In 603 patients 3282 patient days were assessed. Completeness of data entry during the routine phase was 94% for ventilator days and 88% for central venous catheter days. The concordance of infection detection by automated evaluation and evaluation based clinical considerations was fairly good and was quantified by kappa measures of 0.49 for VAP and 0.57 for BSI. Detected infection rates ranged within the German national reference data. Personnel costs for on-ward physicians and infection control personnel were 1.01 Euro per device day in the routine phase. Time expenditure of less than 3 min per device day, rendered in about equal parts by physicians and infection control personnel, was lower than in studies relying on on-ward assessment by infection control personnel.

Hospital electronic medical record–based public health surveillance system deployed during the 2002 Winter Olympic Games

Several computer biosurveillance systems are in place to detect events of public health (PH) significance; however, most lack access to timely and detailed patient-level data and investigation of alerts places a strain on PH resources. Hospital-based infection control professionals led a multi-disciplinary team to develop a computer rule-based system that relies on the patient's electronic medical record. The rules operated on HL7 messages transmitted by clinical computing systems and encompassed a variety of types of patient-level data, including laboratory test ordering and results, radiology ordering and reports, emergency room and outpatient clinic visits, and hospital admissions. Laboratory data were mapped to standard vocabularies, and radiology data were processed using natural language-processing algorithms before the rules were applied to filter for events of PH interest. For each rule, statistical process controls were applied to generate alerts when levels exceeded two standard deviations above the mean. The system was deployed at a large hospital in Salt Lake City during the 2002 Winter Olympic Games, and it was accessed 3 times a day to perform surveillance. Daily reports were provided to local PH agencies after preliminary investigation of the alerts. Of the 24 rules monitored, 9 generated alerts on 11 different occasions. The only significant event of PH interest that was noted during the surveillance period was an increase in influenza during the Games. The positive predictive value of the rules varied with a high value (89%) noted for identification of pneumonia from chest radiograph reports by natural language-processing algorithms. With the assistance of a novel computer-based surveillance system linked to the electronic medical record that uses objective, quantifiable events and access to patient data, infection control practitioners could play a front-line role in biosurveillance and facilitate bidirectional communication with PH agencies.

Automatic Detection of Patients with Nosocomial Infection by a Computer-Based Surveillance System: A Validation Study in a General Hospital

To validate an automated system for the detection of patients with nosocomial infection (NI) in an intensive care unit (ICU). Retrospective analysis of data from the hospital information system. We applied 3 different NI suspicion criteria (positive microbiology reports, antibiotic administration, and diagnosis of clinical infection) and compared the results to those of a prospective NI incidence study done in the ICU during the same period (1999-2002). A 250-bed general hospital in Barcelona, Spain. From April 15, 1999, through June 30, 2002, 1380 patients were admitted to the ICU. Of these, 1043 had an ICU stay of more than 24 hours and were included in the study. At least one NI suspicion criterion was present for 242 patients (23.2%); 2 criteria were present for 184 patients (17.6%); and all 3 criteria were present for 112 (11.7%). Comparison of hospital information system data to the results of the prospective study indicated that the combination of 2 criteria demonstrated the most satisfactory sensitivity (94.3%; 95% confidence interval [CI], 79.5%-99.0%) and specificity (83.6%; 95% CI, 76.8%-88.9%). The positive predictive value was 55.9% (95% CI, 42.5%-68.6%); the negative predictive value was 98.5% (95% CI, 94.2%-99.7%). The system could assign a site of infection for 90.4% of the NIs detected. The hospital information system was a useful tool for retrospectively detecting patients with an NI during the ICU stay. Given its high sensitivity, it may be useful as an alert for the NI team.

Validation of a multicenter computer-based surveillance system for hospital-acquired bloodstream infections in neonatal intensive care departments

Automated systems can facilitate surveillance for health care–associated infections. The New York Antimicrobial Resistance Project (NYARP) electronically monitors trends in bloodstream infections from 6 medical centers in New York, NY. To validate NYARP's data, episodes of health care–associated bloodstream infections detected by this system were compared with those obtained by an infection control practitioner performing an unrelated study in 2 participating neonatal intensive care departments. The sensitivity (84%), specificity (99%), and positive (84%) and negative (99%) predictive values of NYARP were excellent when coagulase-negative staphylococcal bloodstream infections were removed.

Using a Multi-Center, Computer-Based Surveillance System: Overcoming Data Collection Challenges with the Use of Technology and Creative Teamwork

Using a Multi-Center, Computer-Based Surveillance System: Overcoming Data Collection Challenges with the Use of Technology and Creative Teamwork

Using a Multi-Center, Computer-Based Surveillance System: Overcoming Data Collection Challenges with the Use of Technology and Creative Teamwork

ISSUE: Healthcare systems could benefit from a system-wide infection control surveillance database populated with data from component hospitals to assess the individual and collective effects of their infection programs. The challenge: obtain complete, accurate, and consistent data and, at the same time, not distract focus from prevention efforts. PROJECT: Tenet Healthcare, a system comprised of 113 hospitals, sought to develop an efficient data-collection and reporting system that provided outcome data in real time, allowed total hospital infection surveillance, and minimized manual data entry by using existing electronic hospital information systems. These data were automatically derived from the 1) admission, discharge, transfer database; 2) microbiology laboratory database; and 3) pharmacy database. Manual data entry was limited to case managers entering device use and device days and infection control practitioners (ICPs) identifying nosocomial infections. Accuracy of device data was assured by weekly assessment and cross-checking entered data with billing information and using CDC definitions to define nosocomial infections. Some hospitals assigned one individual to enter all device information to ensure complete data capture. ICPs met monthly to evaluate the system. RESULTS: Case Management, Infection Control, Nursing, and corporate managers collaborated to design an improved data collection system named eCARE (Electronic Clinical Analysis and Resource Efficiency). This program will be released by March 2004. New features include risk stratification, patient level “drill down” reports for all nosocomial infections, and comparative rate reports. LESSONS LEARNED: Focusing on three devices (central venous catheters, ventilators, and urinary catheters) made it easier to track device information. The inclusion of nursing management and directors of case management was crucial to successful denominator data collection, especially when measuring outside intensive care units. While hospital-wide data for internal benchmarking is helpful for identifying overall corporate rates, there is a need to stratify data by similar units to identify specific target areas for improvement. There is also a need to confidentially compare hospital-to-hospital data for external benchmarking.

Models of Computer-Based Outbreak Detection

Summary Despite intensive development of computer-based surveillance (CBS), there has been little work on models of such systems. This paper describes a computational model of CBS-output, inputs, and algorithms required to create the output from the inputs. Starting from an assumed requirement that an ideal CBS system be capable of mitigating a bioterroristic outbreak of inhalational Anthrax, the author shows that an output of CBS should be a continuously updated probabilistic threat assessment, which is used to trigger response actions at levels of threat determined by cost-benefit analyses. The paper considers the types of algorithmic transformations required to map from a full spectrum of types of input data to the required output. The set of input data is much broader than previously considered. The specific complementary relationship between the computational model and the evolving architectural model of the National Electronic Disease Surveillance System is discussed.

A national computer-based surveillance system for tuberculosis notification in Singapore

The notification rate of tuberculosis (TB) among residents in Singapore has been declining at a mean rate of 5.6% per annum, from 307 cases per 100,000 population in 1960 to 54 cases per 100,000 population in 1992. A National TB Notification Registry was set up in 1958 using a manual card system, and was captured into a computer database from 1986. To monitor epidemiological trends of TB in Singapore with more speed, versatility and analytical capabilities, a new microcomputer-based surveillance system was developed in 1993. The main software programmes used in this system were DBase IV (version 1.5) and Epi Info (version 5). These versions could use base memory interchangeably and were therefore incorporated into a single application DBase programme. Security features were incorporated into the programme. The TB database was linked to the National HIV Notification Registry to enhance surveillance of combined TB and human immunodeficiency virus infection (HIV). The system was able to track notifications and TB culture results, address letters and analyze data and enabled prompt dissemination of information. The authors believe that this system would enhance surveillance and provide timely information for national TB control programmes. However, the effectiveness of this system is dependent on an established notification structure with notifications for tuberculosis of sufficient completeness.

The Epidemiologic Surveillance Project: A Computer-Based System for Disease Surveillance

The Epidemiologic Surveillance Project (ESP) is a computer-based surveillance system for the rapid transmission and analysis of national morbidity data. These data are transmitted weekly by state health department computers to the Centers for Disease Control (CDC), using a commercial electronic mail facility (MINET). Participating state health departments transmit data through existing computer systems using a standard record and data transmission protocol. State and national data are rapidly analyzed at the CDC, and reports of these analyses are transmitted through MINET to the states. Maps based on these data and other tabular displays of demographic and epidemiologic characteristics of the diseases being tracked are periodically prepared and distributed. Because the efficiency, speed, and uniformity of tabulation of this system are greater than those of systems that use protocol and telephonic reports of case totals, the ESP may be a model for future surveillance systems.