Rothman Index Mobile and Trend is an interactive display of patients' Rothman Index (RI) scores. The RI is a validated, proprietary score of general patient condition, calculated using 26 clinical
inputs, including vital signs, nursing assessments, and selected lab values. The calculation of the RI score relies on a standardized algorithm that aggregates the statistical mortality risk associated with the values of each of the clinical inputs. For the pediatric population, certain inputs4 are age-adjusted, and the resulting RI score is referred to as a pediatric Rothman Index (pRI).
Inputs to the RI Score
The inputs to the RI score are shown in the tables below.
Table 1: List of Inputs to RI Score - Vital Signs
Table 2: List of Inputs to RI Score - Nursing Assessments
Table 3: List of Inputs to RI Score - Lab Values
Table 4: List of Inputs to RI Score - Supported pressure Ulcer Risk Assessments
Legend for Tables 1,2,3, and 4:
Items in italics (heart rhythm, inputs from the Lab Values category) are not required to compute an RI score. Lab values that are available will be used in computing an RI score if and only if a BUN value that is less than 48 hours old is also available.
1: Temperature and the respiratory assessment (or as an alternative configuration cardiac assessment) are required inputs (i.e., these inputs must always be present to compute an RI score). If one of these required inputs is not updated within 15 hours, a new RI score data point will not result.
2: Heart rhythms are mapped to one of 11 rhythms based on similar statistical risk profiles for RI computation. Rhythms such as sinus rhythm and sinus bradycardia are considered low risk and are displayed as "met" in the drill down table. Sinus tachycardia, atrial fibrillation, atrial flutter, heart block, junctional rhythm, paced, ventricular fibrillation, ventricular tachycardia, and asystole are considered elevated risk and are displayed as "not met" in the drill down table. Heart rhythm values more than 48 hours old are not used as in the RI computation.
3: A score will not be calculated if more than two items from both the Vital Signs and Nursing assessments categories are missing.
4: Creatinine, heart rate, respiration rate, and blood pressure are age-adjusted for the pediatric population.
5: Braden may alternatively use any of the supported scales in table 4. Raw scores are transformed from the source assessment type into a corresponding Braden value for Rothman Index calculation.
6: Food/nutrition assessment uses the Nutrition 4-point subscale of either the Braden of the Braden Q as input.
7: The Safety assessment uses the hospital's Fall Risk Score.
8: The weight of the lab values in the Rothman Index computation linearly decreases from maximum weight to zero weight over 48 hours. if new labs come in within 48 hours the 48-hour clock is restarted. The age of the labs is determined by the collected date/time of the Blood Urea Nitrogen (BUN) value.
Each time a documentation variable is updated, a new data point is computed using the new element and the previously documented variables. Documentation entries that are not updated are carried forward indefinitely if the required variables, i.e. respiratory assessment and temperature, have been updated within the last 15 hours. (Carried forward values will be mentioned again in reference to the Drill Down
Table). If either the respiratory assessment or temperature variables is more than 15 hours old, new RI points will not result.
Temperature and respiratory (or cardiac) assessment must always be present to compute the initial RI. If there is a newly admitted patient with no RI score, the end user should ensure that these two elements have been entered into the patient’s health record. Labs are not required to compute the RI, however if results are available, then they will be used if, and only if, a BUN value is available with a collected date/time that is less than 48 hours old. When the BUN age is greater than 48 hours labs are not used in the RI score calculation until a new BUN results. However, labs are always available for viewing in the Drill Down table.
Nursing Assessment Documentation
Nursing assessments are comprehensive, categorized physical and behavioral bodysystem examinations, the totality of which are meant to provide a “head-to-toe” examination of the holistic patient. Though the clinical aim of each nursing assessment is the same across different hospitals, the precise nomenclature and documentation practices may vary by organization.
As shown above in Table 2, apart from the Braden / Braden Q score, nursing assessment inputs to the RI score can be exactly one of two possible values (“met” or “not met”). “Met” generally indicates that the nursing assessment met a hospital normal standard; “Not met” indicates otherwise.
The following processes are used to translate the information in a nursing assessment to the appropriate binary “met” / “not met” input value. A mapping table is created with the input of the customer organization that determines which information in a nursing assessment indicates that the normal standard has been met or not.
- For the Heart Rhythm assessment, the RI uses one of the 11 heart rhythm inputs including: normal sinus rhythm, sinus bradycardia, sinus tachycardia, atrial fibrillation, atrial flutter, heart block, junctional rhythm, paced, ventricular fibrillation, ventricular tachycardia, or asystole. For RI computation purposes, any heart rhythm received that is not one of these eleven is assigned (based on a master mapping table) to one of these values according to similarity in risk. In the RI Mobile drilldown list, low risk heart rhythms (e.g., normal sinus rhythm and sinus bradycardia) are displayed as "met" and heart rhythms associated with elevated risk are displayed as "not met".
- The Food nursing assessment input is based upon the nutrition subcomponent of the Braden or Braden Q score; a nutrition score of 1 (very poor) or 2 (probably inadequate) is deemed "not met".
- The Safety nursing assessment input is based upon numeric fall risk score; a high score is deemed "not met".
- For the remaining nine nursing assessments:
- When a hospital conducts "charting by exception" nursing assessments (ie., a master question for each assessment determines whether a normal standard has been met), the input to the RI score will be mapped to "met" or "not met" in accordance with the answer to the master question(s).
- When a hospital dose not conduct "charting exception" nursing assessments, a series of sub-questions are collectively assessed for each assessment to determine if there is any deviation from a normal condition, in which case the assessment would be deemed "not met". The following table provides an example of such mapping:
Table 5: Example Portion of a Nursing Assessment Mapping Table
As part of the implementation process, Spacelabs will work with the customer organization to identify the nursing assessment documentation that must be included in the data feed, as well as to ensure that each field is mapped correctly to a “met” or “not met” value for the relevant body-system assessment.
It is important to remember that while the nomenclature and documentation practices between hospitals may vary, the overall clinical information conveyed by each nursing assessment, across different hospitals, is the same (e.g., there is little variability between a "normal" cardiac or respiratory assessment across different hospitals, despite the potential variation in nomenclature or documentation practices between those hospitals).