 |
Research Use of Electronic Patient Information:
|
||
Scenario 1
|
These scenarios represent a series of typical requests for access to patient information from the online clinical information system. The purpose of the scenarios is to test the scope and applicability of the proposed policy for research use of online patient information. The questions below should be asked after each scenario. If the answer to question 1 is ‘yes’, then the scenario falls under the scope of the policy. If the answer to question 1 is ‘no’, then the request for access to electronic patient information is covered by other institutional policies concerning access to patient information. The remaining questions help to clarify what additional steps (beyond RSRB review) may be needed.
The proposed policy addresses circumstances under which RSRB review is required for the use of electronic patient information. There are additional considerations to be addressed in terms of procedure. These questions should also be considered by the investigator if they affect the implementation:
A couple of years ago, I read about a study showing that 45% of patients with stable treated heart failure may have episodes of apnea and hypopnea during sleep. I wondered if some kind of laboratory test might help identify patients at risk for this problem, so we could intervene before these problems occur. Arterial PCO2 has an influence on breathing. I decided that the first step was to look at the hospital’s patient database, to find patients with a history of heart failure and see if an association between levels of PaC02 and sleep problems could be confirmed. My strategy is to request patient records with any discharge diagnosis that included heart failure, insomnia and apnea. Then, I will stratify PaCO2 data into various groups (to be determined). If I find a trend in the data, I will design a trial of a diagnostic test that can identify cardiac patients at risk for apnea. I'm interested in how (or whether) collaboration between nurses and physicians has an impact on ICU patient outcomes. I need to collect data from providers about collaboration, but I also want to get APACHE III scores for ICU patients for their first 24 hours in the ICU. For each patient, I need to know primary ICU diagnosis, location prior to ICU admission and duration at that location, age, presence or absence of certain co-morbidities (AIDS, hepatic failure, lymphoma, metastatic cancer, leukemia/multiple myeloma, immune suppression, cirrhosis, chronic dialysis), and, if a surgical admission, whether surgery was elective. From the first 24 hours of ICU stay, I need vital signs, whether the patient was on a ventilator, blood gases, selected laboratory values, total urine output, and Glasgow coma scores. My colleagues on the in-patient cardiology unit have been complaining to me that it’s impossible to keep up with the increasing demands for written documentation and physician approval of care procedures recorded in the patient’s chart. I think computing technology can help reduce the burden. I want to create an experimental electronic record system, one that will be used as the physician’s primary documentation device. It will be used to review prior visits and to document the current one. It will permit the integration of voice and picture data as well as text and check box entries. The program will use medically relevant data to provide the classifications needed for billing. To assess both the values and the obstacles to this kind of approach, the prototype system needs to be tested in real life situations. To build and test the prototype, I need to build a database consisting of all relevant patient data, including laboratory results, for each admitted patient of each physician participating in the project. I’m an internal medicine resident at Strong Memorial Hospital. We have a clinical information system, but I have to go to the floor unit to find a workstation where I can send e-mail to someone about a patient, or check lab results or order a procedure. The competition for access to the floor unit computer is pretty intense, so I don’t really get time to think when I’m using it. I heard at talk at a meeting last month about a place where physicians are using hand-held computers to enter orders and check results. This frees them from having to go to the floor unit workstation to get their work done. I talked with someone in medical informatics who has a wireless pen-based tablet. We want to try a test. The tablet would run the standard client software for the clinical information system and would interact with that system directly. We would give tablets to all the house staff on my service and let them do orders and documentation from either the tablet or the floor unit workstation. Informatics faculty would monitor how quickly (and from where) orders are using the tablet as opposed to the fixed terminal environment. I’m a nurse in the Visiting Nurse Service, providing follow-up care to patients from Strong Memorial Hospital. I carry a laptop on my visits, and record my notes and observations there. The laptop is heavy, and I still have to enter information into the patient record in the clinical information system after each visit. I talked with someone in medical informatics who has a small hand-held computer. We want to try a test. The hand-held would run special software that would present the patient record, and would allow me to add data and notes. Before a visit, informatics would download my patient’s files from the clinical information system and reformat them for my hand-held. After a visit, they would reformat and upload them into the clinical information system. In our test, we would give hand-helds to half the visiting nurses and let the others continue with the existing process. Informatics would do data reformatting, and would compare the speed, accuracy and user satisfaction of the two approaches. I am a faculty member in Pharmacology & Physiology in the School of Medicine & Dentistry. I want to do a study on comparative effects of a nutritional intervention for juvenile and adult-onset diabetes. I want to search the institution’s records for all patients that have diabetes, are between the ages of 10 and 18 or 35 and 50, and are within normal boundaries for weight. Once I find a study cohort, I will contact those patients and invite them to participate in my study. I’m a primary care physician working in a community office. I have admitting privileges at Strong Memorial and Highland. About 50% of my patients are covered for care through a managed care contract I have with StrongCare. I have the feeling that my managed care patients are using health care services more frequently than my other patients. If I find that this is the case, I will adopt a more aggressive patient education program. I want to compare the frequency of care events for my managed care patients with that of my other patients. For 100 patients in each category, I need records for all encounters during the past year. I need information on ambulatory, in-patient, emergency room, home care, and nursing home visits. A new inexpensive procedure for measuring hemoglobin types has appeared in the literature. A researcher in Pathology/Laboratory Medicine wants to evaluate this new technique. He can use extra blood samples to be discarded for this experiment, but he must identify appropriate blood samples first, so will take those which have already been analyzed using chromatography. Besides the abnormal Hb AS, SC, SS, thalassemias, etc., he needs normal controls. He will randomly pick blood samples and run the chromatographic analysis on them at his own expense. He must do a quick, cursory scan of each patient's clinical information (just the diagnosis) to make sure these individuals have no other clinical condition that will invalidate the sample's use in this experiment. He has access to patient records in the Laboratory medicine system, but he wants to use the patient records in the clinical information system for the additional information he needs to identify appropriate samples. Based on his own in vitro work and on the literature, an immunologist postulates that a specific type of T-cell, identifiable with a newly developed monoclonal antibody, in maternal blood during pregnancy can serve as an early indicator of premature delivery. He needs blood samples from pregnant women to develop this assay. He obtains these anonymously from the clinical lab in the leftover sample. Having completed his preliminary assays, he is considering going forward to a full-scale clinical study. To make this decision, he needs to compare blood samples of women needing a cesarean with those delivering vaginally. He can still use leftover blood samples from the lab, but to be sure the comparison was valid, he needs to review the patient’s chart for each sample used. The researcher is a pediatrician, with admitting privileges at Strong Memorial Hospital; he has access to the CIS and the charts for his pediatric patients. A researcher is preparing a grant proposal to send to NIH. The proposed project evaluates cancer markers for progression and prognostic information in esophageal cancer. The researcher proposes to obtain samples from patients who visit the Cancer Center on a regular basis for ongoing follow-up. The attending will take a swab of the patient’s throat and prepares a standard slide for cytopathology. He will then place the swab (which would normally have been discarded) in a special vial of preservative which the researcher has provided. The researcher will use this to extract the cellular material she needs. The proposed approach conforms with the consent for research use of tissues and organs gained from patients admitted to Strong Memorial Hospital which states: "I authorize Strong Memorial Hospital to dispose of any tissues or organs removed as a result of the procedure authorized above, or to preserve such tissues or organs, in its discretion, for scientific or teaching purposes." However, to actually carry out this research, the researcher needs additional information which can easily be obtained from the clinical information system.
|
||
 |
Back to URMC Electronic Patient Information Policy Statement | ||
 |
Appendix 1: Policies Relating to Human Subjects Research | ||
 |
Appendix 2: Rationale and Context for the URMC Policy | ||
 |
|||