Clark, J.A., and Talcott, J.A. (2001). "Symptom indexes to assess outcomes of treatment for early prostate cancer." Medical Care 39(10), pp. 1118-1130.
Both radical prostatectomy and external beam radiotherapy treatments for early prostate cancer often cause physical side effects, including urinary, bowel, and sexual dysfunction, which change the quality of men's lives. These researchers prospectively evaluated the outcomes of nearly 200 patients undergoing one or the other treatment by asking them to complete self-administered questionnaires before treatment and 3 and 12 months after treatment. They developed indexes of urinary, bowel, and sexual function and symptom-related distress based on questionnaire responses. Symptom and symptom-related distress indexes in each domain were highly correlated. The indexes accounted for significant proportions of the variance in health-related quality of life measures for these patients. The researchers conclude that these indexes may be helpful in monitoring outcomes of treatment for early prostate cancer.
DiSalvo, T.G., Normand, S.T., Hauptman, P.H., and others. (2001, September). "Pitfalls in assessing the quality of care for patients with cardiovascular disease." American Journal of Medicine 111, pp. 297-303.
There are no clinical performance measures for cardiovascular disease that span the continuum of care from the hospital through postdischarge ambulatory care. After reviewing practice guidelines and the medical literature, these investigators developed potential performance measures related to cardiovascular disease therapy, diagnostic evaluation, and communication. They tested the feasibility of developing and implementing such measures for 518 patients with heart attack, 396 with congestive heart failure, and 601 with hypertension, who were enrolled in four major U.S. managed care plans at six geographic sites. They found that constructing meaningful clinical performance measures was straightforward, but implementing them on a large scale would require improved data systems. For example, diagnosis at discharge often didn't match administrative and records data, medical records were missing, and there were problems in identifying physicians accountable for care. In addition, many cases were excluded from measures of appropriate therapy because the measures were conditional on test results, and rates of performing key diagnostic tests were low.
Espino, J.U., and Wagner, M.M. (2001). "Accuracy of ICD-9-coded chief complaints and diagnoses for the detection of acute respiratory illness," in Proceedings of the AMIA Annual Fall Symposium, 2001. Philadelphia: Hanley & Belfus; pp. 164-168.
When clinicians fail to notice or report bioterrorist or naturally occurring disease outbreaks, public health systems are the next line of defense. Chief complaints and diagnoses from emergency departments (EDs) that are coded using the ICD-9 (International Classification of Diseases, Ninth Revision) and routinely collected for electronic submission of insurance claims have potential for use in public health surveillance, according to these authors. They constructed two detectors of acute respiratory illness: one based on ICD-9-coded chief complaints and one based on ICD-9-coded diagnoses, whose performance they measured against the human classification of cases based on review of ED reports. Using ICD-9-coded chief complaints, the sensitivity of detection of acute respiratory illness was 0.44 and its specificity was 0.97. The sensitivity and specificity using ICD-9-coded diagnoses were no different. These findings, coupled with the timeliness and electronic availability of such data, support use of detectors based on ICD-9-coding of ED chief complaints in public health surveillance.
Frosch, D., Porzsolt, F., Heicappell, R., and others (2001). "Comparison of German language versions of the QWB-SA and SF-36 evaluating outcomes for patients with prostate disease." Quality of Life Research 10, pp. 165-173.
The Quality of Well-Being Scale (QWB) and Medical Outcome Study 36-item short form (SF-36) are two different methods for measuring general health outcomes. Few studies have compared these approaches with one another, and no studies have compared German-language versions. These researchers administered the German QWB-SA and a German-language version of the SF-36 to clinical population groups with current diagnoses of prostate cancer, benign hyperplasia of the prostate, colon cancer, and rectal cancer. The researchers obtained data from German clinics on quality of life measures, cancer stage, and disease state. The QWB-SA and SF-36 were highly correlated. The QWB-SA was systematically related to disease state. Those with no symptomatic evidence had the highest scores followed by those who were stable with no metastatic disease and those with metastatic progression. Similar patterns were found for most SF-36 scales. However, the SF-36 did not discriminate between those with no evidence of disease and those with stable disease without metastasis.
Jenders, R.A., and Shah, A. (2001). "Challenges in using Arden Syntax for computer-based nosocomial infection surveillance," in Proceedings of the AMIA Annual Fall Symposium, 2001. Philadelphia: Hanley & Belfus; pp. 289-293.
The average incidence of nosocomial (hospital-acquired) infection (NI) is 5 to 10 percent, sometimes reaching 28 percent in intensive care units. Detection of NI outbreaks typically requires daily manual review of microbiology laboratory test results, which is prone to error and may miss trends in infection. In order to facilitate the computer-based detection of NIs, these investigators created a two-phase system. The first phase uses Arden Syntax to filter microbiology laboratory data in order to retain only those results suggesting actual infection. The second phase compensates for the single-patient focus of most installations of Arden Syntax by using a statistical monitor to track results over many patients across multiple hospital inpatient units. Preliminary data suggest that the first phase provides a significant reduction in the volume of messages that must be processed. The authors conclude by suggesting improvements in the Arden Syntax that would facilitate detection of NIs.
Lobach, D.F., Low, R., Arbanas, J.A., and others. "Defining and supporting the diverse information needs of community-based care using the Web and hand-held devices," in Proceedings of the AMIA Annual Fall Symposium, 2001. Philadelphia: Hanley & Belfus; pp. 398-402.
Economic factors are shifting the focus of care from the hospital to the community. Community-oriented initiatives, however, often require partnerships that cross traditional boundaries. As a result, the initiatives often lack a common information infrastructure to support the care delivery process. These authors created and implemented a Web-based information and communication system to support the needs of a community-based healthcare project for Medicaid beneficiaries in Durham County, NC. They identified the relevant information requirements and stakeholders for community-based care and created a system interface that required only a Web browser and an information distribution system that used electronic mail. They also explored the use of hand-held devices by providers to download information from a clinical database and to access and collect patient information at the point of contact. The overall goal of the project was to lower costs and improve the quality of community-based health care through improved handling of information.
Ortiz, E., Meyer, G., and Burstin, H. (2001) "The role of clinical informatics in the Agency for Healthcare Research and Quality's efforts to improve patient safety." in Proceedings of the AMIA Annual Fall Symposium, 2001. Philadelphia: Hanley & Belfus; pp. 508-512.
The Institute of Medicine (IOM) issued a report on medical errors in 1998, which estimated that up to 98,000 people die in U.S. hospitals each year from errors. This report raised concerns about patient safety and suggested that this public health problem should be addressed like other epidemics such as heart disease, diabetes, and obesity. In 2001, the IOM released a followup report encompassing a broader range of quality issues. It concluded that the U.S. health care system is outmoded and incapable of providing consistent, high-quality care. The report also outlined a strategy for redesigning U.S. healthcare delivery to achieve safe, dependable, high-quality care, which emphasizes information technology as an integral part of the solution. The Agency for Healthcare Research and Quality is making a substantial investment in initiatives to reduce medical errors and improve patient safety. AHRQ developed a series of research solicitations that form an integrated set of activities to design and test best practices for reducing errors in multiple health care settings. This paper discusses the components of the program and the central role of medical informatics research in the Agency's efforts to improve patient safety in America.
Patrick, D.L., Engelberg, R.A., and Curtis, J.R. (2001, September). "Evaluating the quality of dying and death." Journal of Pain and Symptom Management 22(3), pp. 717-726.
Improving the quality of end-of-life care is a priority for patients, families, and clinicians. These authors propose a model to evaluate the quality of dying and death based on concepts elicited from literature review, interviews with people with and without chronic and terminal conditions, and consideration of desirable measurement properties. They defined the quality of dying and death as the degree to which a person's preferences for dying and the moment of death agree with observations of how the person actually died, as reported by others. They modified expected level of agreement by circumstances surrounding death that may prevent following a patient's prior preferences. The researchers derived six conceptual domains (symptoms and personal care, preparation for death, moment of death, family, treatment preferences, and whole person concerns) that encompassed 31 aspects of care. These could be rated by patients and others as to their importance prior to death and assessed by significant others or clinicians after death to assess the quality of the dying experience.
Peleg, M., Ogunyemi, O., Tu, S., and others (2001). "Using features of Arden Syntax with object-oriented medical data models for guideline modeling," in Proceedings of the AMIA Annual Fall Symposium, 2001. Philadelphia: Hanley & Belfus; pp. 523-527.
Computer-interpretable guidelines (CIGs) that are linked to electronic medical records (EMRs) can provide patient-specific advice automatically at the point of care. There are several methods for
encoding guidelines to make them computer-interpretable. All of these methods have constructs for defining criteria that relate medical concepts to patient data. Although each method has different constructs, they all use some sort of expression language for specifying local decision and eligibility criteria and a data model for medical concepts and patient data. These investigators describe how they used features of Arden Syntax with object-oriented medical data models for guideline modeling.
Sanders, G.D., Nease, Jr., R.F., and Owens, D.K. (2001). "Publishing Web-based guidelines using interactive decision models." Journal of Evaluation in Clinical Practice 7, pp. 175-189.
These investigators developed a Web-based system, ALCHEMIST, that automatically creates evidence-based guidelines which can be disseminated, tailored, and updated over the Web. They demonstrated the use of the ALCHEMIST system to develop Web-based guidelines for three clinical scenarios: chlamydia screening for adolescent women, antiarrhythmic therapy for the prevention of sudden cardiac death, and genetic testing for the BRCA breast cancer mutation. Using ALCHEMIST, they demonstrated that tailoring a guideline for a population at high-risk for chlamydia changes the recommended policy for control of the infection from contact tracing of reported cases to a population. They used ALCHEMIST to incorporate new evidence about the effectiveness of implantable cardioverter defibrillators (ICD) and demonstrated that the cost-effectiveness of ICD use improved from $74,000 per quality-adjusted life year (QALY) gained to $34,500 per QALY gained. Finally, they showed how a clinician could use ALCHEMIST to incorporate a woman's preferences for various health states to develop patient-specific recommendations for BRCA testing, which improved quality-adjusted life expectancy by 37 days.
Schneeweiss, S., Seeger, J.D., Maclure, M., and others (2001). "Performance of comorbidity scores to control for confounding in epidemiologic studies using claims data." American Journal of Epidemiology 154, pp. 854-864.
Coexisting illnesses (comorbidity) with the one being studied is an important confounder in epidemiologic studies. These authors compared the predictive performance of comorbidity scores for use in epidemiologic research with administrative databases. The study participants were elderly Canadians who received angiotensin-converting enzyme inhibitors or calcium channel blockers at least once during the observation period. The researchers computed six scores for all 141,161 participants during the baseline year (1995-1996). Endpoints were death and health care use during a 12-month followup (1996-1997). Four scores based on the International Classification of Diseases, Ninth Revision (ICD-9) generally performed better at predicting 1-year mortality than the medication-based Chronic Disease Score (CDS)-1 and CDS-2. Number of distinct medications used was the best predictor of future physician visits and expenditures and a good predictor of mortality.
Shekelle, P.G., Park, R.E., Kahan, J.P., and others (2001). "Sensitivity and specificity of the RAND/UCLA appropriateness method to identify the overuse and underuse of coronary revascularization and hysterectomy." Journal of Clinical Epidemiology 54, pp. 1004-1010.
The RAND/UCLA appropriateness method, which combines expert opinion with scientific evidence, has been used frequently in the United States and other countries to assess the appropriateness of medical procedures. It has been criticized for being potentially sensitive to panelist selection and potentially susceptible to misclassification (that is, labeling a procedure "inappropriate" when it was "appropriate" and vice versa). These researchers performed a parallel three-way replication of the appropriateness panel process for each of two procedures, coronary revascularization and hysterectomy. They demonstrated that the sensitivity and specificity of this method for identifying the overuse and underuse of coronary revascularization and the overuse of hysterectomy were comparable to the sensitivity and specificity of commonly used diagnostic tests. However, they cautioned that the imperfection of this method can lead to a clinically significant misclassification bias.
Tsui, F-C., Wagner, M., Datao, V., and Chang, C-C. (2001). "Value of ICD-9 coded chief complaints for detection of epidemics," in Proceedings of the AMIA Annual Fall Symposium, 2001. Philadelphia: Hanley & Belfus, pp. 711-715.
The threat of bioterrorism has elevated the importance of improving the Nation's capability to detect epidemics. These researchers assessed the usefulness for early detection of epidemics of
chief complaints, coded using the ICD-9 (International Classification of Diseases, Ninth Revision), at the time of a patient's arrival at the emergency department. The authors measured sensitivity, positive predictive value, and timeliness of influenza detection using a respiratory set (RS) of ICD-9 codes and an influenza set (IS). They also measured inherent timeliness of these data using the cross-correlation function. For a 1-year period, the detectors had a sensitivity of 100 percent (1/1 epidemic) and positive predictive values of 50 percent for RS and 25 percent for IS. The timeliness of detection using ICD-9-coded chief complaints was 1 week earlier than the detection using pneumonia and influenza deaths (the gold standard). The inherent timeliness of ICD-9 data measured by the cross-correlation function was 2 weeks earlier than the gold standard.
Zeng, X., and Wagner, M. (2001). "Modeling the effects of epidemics on routinely collected data," in Proceedings of the AMIA Annual Fall Symposium, 2001. Philadelphia: Hanley & Belfus; pp. 781-785.
The development of computerized epidemic early detection systems has stimulated interest in new approaches to public health surveillance based on analysis of routinely collected data. The key underlying this new paradigm is that epidemics perturb the normal patterns of over-the-counter drug purchases; work and school absenteeism; emergency room visits; and other routinely collected data. These authors reviewed behavioral and cognitive models of patients' responses for diseases that would cause symptoms similar to those caused by known bioterrorism agents. They combined ideas from these models with a model of early detection of bioterrorism attack from routinely collected data. They conducted a literature review on factors influencing patients' behaviors and the pattern of health service use after onset of symptoms such as shortness of breath, which would conceivably be a result of diseases caused by bioterrorism attacks. The study focused on human behavior, such as care seeking and information seeking, in the period between the onset of initial symptoms and the first visit to health care facilities. The goal was to build a model relating known factors about these behaviors and their effects on routinely collected data, which may be useful to researchers in early bioterrorism detection, simulation, and response policy analysis.
Return to Contents
Current as of January 2002
AHRQ Publication No. 02-0012
Internet Citation:
Research Activities newsletter. January 2002, No. 257. AHRQ Publication No. 02-0012. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/research/jan02/
540 Gaither Road Rockville, MD 20850