The investigators propose a randomized control trial performed on the inpatient medical service of Wishard Memorial Hospital. This hospital has 20 intensive care beds and two 40 bed wards. The general internal medical service is comprised of eight discrete services, each attended by a team house staff, students and attending faculty, all of whom serve one month rotations. The physician interventions will randomly divide the eight ward services into two groups of four services each. Four teams will be randomly assigned into the intervention group and will receive computer based messages when writing orders on the work station. The physicians in the control group will also write orders on the work station, but will not receive the computer generated messages for the study. They will, however, continue to receive the other computer-generated suggestions that have been designed for other protocols such as the drug interactions and negative detailing orders that are part of the operating system. The investigators have previously used the proposed study design to test other computer-based interventions in their institution. The nurses will be divided into intervention and control groups based on two wards in which they participate, A or B. Only the intervention nurses will receive the computer reminders. The computer assisted intervention has two parts: 1) a patient centered data base, and 2) the event triggered computer reminder system. The patient centered data base will contain patient name, address, phone number, information on household composition, financial resources, the patient's primary care physician, patient's self-reported health and functional status, self-reported compliance, knowledge of chronic disease and self management, health risk behaviors and educational level. The investigators provided examples of the types of event-triggered reminders that would be in the system. For instance, when the physician enters admitting orders, they would get a reminder stating that """"""""36 percent of the patients reported that there was no doctor in charge of their care while they were in the hospital. Upon admission and repeatedly throughout Mrs. Jones stay, please clearly identify all members of your team and their roles."""""""" The patient survey will then include questions specifically linked to this reminder such as: How would you rate how well all of your doctors work together? and; Was there one doctor who was in charge of your care while you were in the hospital? Additional examples are provided on pages 73-75. Patients will be contacted while in the hospital in order to receive informed consent to participate in the study. A detailed protocol is provided about how patients will be contacted. In order to increase the response rate, the investigators will use both mailed questionnaires and telephone interviews. As in the first application, secondary patient outcomes include health care resource utilization, measured by the number of hospitalizations and ER visits within three months following discharge from the hospital, as well as the number of patients keeping post discharge follow-up appointments. Additional secondary outcomes also includes subjective health status. Covariates that will be measured include patient demographic characteristics, insurance status, discharge diagnosis and comorbid conditions (including asthma, chronic obstructive pulmonary disease, congestive heart disease and ischemic heart disease). Measurements include the specific Activity Scale, the Symptom Response Ratio and the Charlson Comorbidity Index. Patients will be the unit of analysis. The investigators plan to account for potential interactions between physicians and nurses within each intervention, as well as patients clustered with clinical teams. Methods to be used to handle clustered data were not specified. The investigators plan to use generalized linear models, in conjunction with generalized estimating equations to model patient satisfaction. They also plan to account for possible biases induced by different modes of survey administration.
