Remote Post-Operative Rounding and the Physician Ability to Identify Morbidity
Remote Post-Operative Rounding and the Physician Ability to Identify Morbidity
Telerounding is the use of wireless remote video-confrencing to assess hospitalized patients. Physicians thus rely on all ususal data collected during bedside rounds with the exeption of a direct physical exam. The intention of this study is to determine if post-operative morbidity can be identified in an accurate and timely manner.
Design: Two hundred seventy patients were recruited from three sites into a prospective multi-center randomized trial. Participating institutions were the University of California, Davis Medical Center (Sacramento CA), Johns Hopkins Hospital (Baltimore MD) and Sentara Health (Norfolk VA). Internal Review Board approval was granted at each institution independently. The study population consisted of patients scheduled for urologic surgery, over age 18, and able to read and understand English. Patients undergoing the following laparoscopic procedures were offered participation: nephrectomy, partial nephrectomy, nephroureterectomy, retroperitoneal lymph node dissection, partial ureterectomy, and radical prostatectomy. Patients had an expected hospital stay of 24 to 72 hours. Eligible patients who were unable to provide consent or who do not wish to participate in the study received the standard of care provided at each institution. A total of two hundred seventy patients consented to participation.
Randomization: A stratified block scheme was used for randomization. Participating patients were randomized to either: standard daily bedside rounding by the attending surgeon (control arm), or daily telerounding only by the attending surgeon (intervention arm).
Outcome Measures: The primary outcome measure was differences in rates of attending surgeon-identified complications between standard rounds and telerounds. The secondary outcome measures included differences in length of stay and patient reported satisfaction with their hospitalization.
Protocol: Consenting patients had their scheduled surgery and received the standard peri-operative and immediate post-operative care. All patients were managed with a rapid recovery protocol. This included: liquid diet beginning twelve hours after surgery with immediate advancement as tolerated; complete blood count, serum electrolytes, BUN and creatnine in the recovery room and each morning until discharge; and usual nursing data recorded each shift. These data points included oral temperature, blood pressure, pulse, respiratory rate, fluid intake, fluid output, and pain scale Once transferred to the patient floor, all patients communicated with their attending physician on a daily basis. The visit, either at the bedside or via telerounds followed a set scripting. The visit was conducted between the patient and the attending without other staff present. The focus of the visit was review of objective data (vital signs, fluid balances and laboratory values), subjective data (cursory abdominal exam if at the bedside and evaluation of drain effluent) and a discussion of the anticipated goals for the day. Visit duration was timed. Intervention patients could remove themselves from the study at any time by requesting a bedside visit by the attending physician. Telerounding concluded with either the hospital discharge or identification of a major post-operative complication.
Definitions: A major complication was defined as an event that required transfer to a monitored setting. Minor complications included events that delay discharge greater than 24 hours beyond the expected length of stay. For example (but not limited to): post-operative ileus, drop in hematocrit, prolonged drain output or fever Event monitoring: Identification of complications was recorded prospectively. Usual resident-level bedside rounds were maintained throughout the study. The resident team and the attending surgeon recorded identified events independently, thus allowing for evaluation of concordance. This dual rounding design served as a minimally acceptable standard as stipulated by the various internal review boards. As a precondition of IRB approval, identification of an event required notification of the attending surgeon in a timely fashion.
Instrument: We utilized a validated 21-item questionnaire to evaluate patient ratings of their hospital care. Items regarding postoperative care were designed using an extensively tested and validated response scale (1, poor to 5, excellent). Item stems were modified from Patient Judgments of Hospital Quality to make them more salient to the postoperative experience. Five items asked patients to rate their baseline health status as well as their health status during the hospitalization. Seven items asked patients to evaluate aspects of the care they received while an in-patient. Nine items asked those randomized to the telerounding arm to evaluate the telecommunications system and to indicate their level of interest in having this system incorporated into usual post-operative care.
Device: The telerounding robot is a 60-inch tall wheel driven device. The robot consists of the motor base unit, Pentium III CPU, high definition digital camera, flat screen monitor and microphone. Data to and from the robot is transferred over a high-speed wireless network and integrated with proprietary software. The physician connects remotely to the robot via a base station. The base station consists of a Pentium III desktop computer, high definition digital camera, flat screen monitor, microphone and joystick controller. Each of the institutions used identical technology. Prior testing of the system demonstrated imperceptible video and audio delay.
Analytic Plan: The primary endpoint of the study is patient morbidity. The expected rate of complication (major and minor) after a laparoscopic urologic procedure is 16%. Based on our power calculation, a total of 270 patients (135 in each arm) were required to detect a 1% difference in complication at the 0.05 alpha level and the 0.8 beta level. Continuous variables were compared using student T-test. Proportions were compared using chi-square analysis. Logistic regression was reserved for adjustments based on observed statistical differences in baseline demographic data.
Inclusion Criteria:
Exclusion Criteria: