2.50
Hdl Handle:
http://hdl.handle.net/10755/335055
Category:
Full-text
Type:
Presentation
Title:
Exploring Nursing Cost Using Patient Level Data
Other Titles:
Exploring Technology to Improve Patient Care
Author(s):
Jenkins, Peggy A.
Lead Author STTI Affiliation:
Alpha Kappa
Author Details:
Peggy A. Jenkins, PhD, RN, pjenkins0701@gmail.com
Abstract:
Session presented on Sunday, July 27, 2014: Purpose: A focus in the American healthcare system is on value-based delivery of services. Payment models will reward hospitals that can efficiently and effectively care for patients. Since nursing is the largest labor segment in the hospital, understanding this very important input is imperative in delivering high quality care at an affordable price. Traditional methods of accounting for nursing services have included measuring nursing hours per patient day, which is a unit level measurement that does not include variability in nursing care at the patient level. Nursing administrators, healthcare leaders, payers, and patients should be interested in understanding the cost of nursing service at the patient level. To benchmark patient level costs, hospitals need to move beyond current unit level measurements of cost to analytics linking individual patients to nurses. Using patient level data in nursing research provides a source capable of answering many unrequited questions about individual nursing contributions to patient outcomes. The current science of nursing cost will evolve and evidence can be provided through which value based healthcare is better built across the world. 1,2,3,4,5,6,7,8 The purpose of the study was 1) to explore the variability of nursing cost per acute care episode for patients with similar DRGs with and without major complications; 2) to investigate the relationship among patient characteristics, nurse characteristics, nursing intensity, and nursing cost as a patient outcome. Four research questions included: What is the variation in nursing cost per acute care episode for patients with the same DRGs without complications, with complications, with major complications? What is the relationship among nursing characteristics (age, years of service, educational degree) on nursing cost per acute care episode? What are the characteristics of nurses assigned to patients with complications and major complications? What is the average nursing cost per day measured at the patient level? Methods: A retrospective, exploratory, cost study using secondary patient level de-identified data was completed. The study site was one general medical surgical unit in a large academic medical center located in the Central United States. The study site organization has earned Magnet designation. Sample was 3111 patients and 150 nurses on the unit over a two-year period. Important and time-consuming steps in secondary data analysis included data acquisition and management. The collaborative data acquisition process was complex and took almost two years to accomplish. The source of data was Clairvia Care Value Management Patient Assignment software plus medical management and human resources databases. De-identification of data was completed before the researcher obtained data. Staff time to de-identify data was negotiated via conference calls and on-site meetings by the researcher with the president of the software company and the nurse scientist and other significant staff from the study site. The researcher used a model for data management consisting of four steps; 1) planning, 2) organization, 3) computing, 4) documentation. 9 Stata do-files provided a tool to systematically record commands for each step of the data management and analysis process. Forty-five variables were collected, cleansed, and new variables were constructed for a total of forty-nine final study variables. Data were analyzed at the shift level, day level, and patient episode of care level according to the research question. Shift level patient and nursing intensity data were one original source of data, which were aggregated to patient episode of care to understand nursing cost per acute episode. Patient and nurse characteristic data were another source of data that were merged with shift level data. The conceptual definition of nursing cost per acute care episode (NCACE) was comprised of lower level elements (NI-nursing intensity) that were aggregated to form the emergent model (summation of individual nurse intensity per patient multiplied by individual nurse hourly wage). NCACE represents Nursing Cost per Acute Care Episode, which was defined as the summation of the product of nursing intensity (NI) and nursing hourly wage (NW).

NI= f (time spent with patient) + (skill level of nurse) Nursing intensity was a function of time spent with the patient measured using Clairvia demand-driven patient assignment software. Nursing intensity was calculated in the patient assignment software based on the following methodology. The acuity score was a 1-12 point scale derived from an outcomes-driven model grounded in the Pesut and Herman conceptual framework, the Outcome-Present State-Test Model of Clinical Reasoning. 10 Nurses rated patients every shift or when condition changed on several outcomes using a 1-5 point scale that contributes to an algorithm producing the nursing intensity score. A monthly audit was completed to assure interrater reliability of acuity measurement and the result was 86% accuracy. Skill level of the nurse was recorded in the Clairvia software as RN, LPN, and Patient Care Associate. NW= f (experience, years of service, education, certification) Nursing wage was operationalized using actual hourly wage for each nurse providing direct care. The principle diagnosis was measured through the DRG. Complications were measured using a four level variable 'compcode'. The four value labels of the variable were 0 = no drg, 1= without complications, 2 = with complications, 3 = with major complications. Using the same DRG without, with, and with major complications allowed for a comparison of nurses assigned to increasingly complex patients. Nurse characteristics were measured and analyzed in relationship to cost of nursing care. Variables describing the nursing unit that are contextual included unit type, number of beds, average acuity per patient, unit skill mix. Nurse characteristics measured included skill level, age, education level, years of service at institution, years of service on the unit, and float. De-identified data from three databases were merged into a single file and analyzed using Stata software. Correlation analysis and regression analysis were used to explore relationships among patient characteristics, nursing characteristics, and nursing cost per acute care episode. Microeconometric measurement was used to determine the elasticity of nursing characteristics on patient acuity and direct nursing cost per patient. IRB expedited approval and continuation was obtained from the study site and the researcher's University. Results: Key findings included 1) patients with the same diagnosis have large variability in nursing intensity and nursing cost by shift, day and acute care episode (i.e. cost per day range DRG 192 $5.68-287.37, 191 $5.96-257.56, 190 $10.06-366.86); 2) nurses may not be assigned patients based on experience and education level; 3) direct nursing cost per patient on the study unit was $96.48 on average per day, which was only 5.8-7.3% of the daily room and board charge. Conclusion: There is large variability in direct NCACE for patients with similar DRGs. An example is patients with COPD without complications (NCACE range $54-1570, M $325, SD $242); COPD with complications (NCACE range $17-3674, M 408, SD $427); COPD with major complications (NCACE range $132-1455, M $462, SD $316). Nurse scientists have provided evidence for variability in nursing cost for patients with similar DRGs for decades, yet hospitals in America continue to be reimbursed under an assumption that patients with similar DRGs receive the same amount of nursing care. 11,12,13,14 This study refutes the assumption. RN years experience in the organization was the nurse characteristic most associated with direct nursing cost. A 10% increase or 9.3 total nurse years experience in the organization for the patient episode of care is associated with a 9.9% or $34.92 increase in direct cost of nursing care per episode for patients on the study unit holding all other variables constant. Data did not support the hypothesis that nurses with greater experience or education level are assigned sicker patients. Average RN experience assigned per patient episode was not significant when regressed on average patient acuity. Percent of BSN nurses assigned was significant in the model with nominal effect. The mean nursing direct cost per day for all patients in the study was $96.48 ($55.73, range-$.33-600.81). The room and board charge for each patient in a medical/surgical unit at the study hospital ranges from $1321-1650 per day. Therefore, the direct nursing cost per day is only 5.8-7.3% of the daily room and board charge. Direct nursing care is a small percent of the cost, but patients don't know this because direct nursing cost is included in the room and board charge and not itemized on the patient bill. Limitations of the study include de-identified data from a secondary source were used and cases with missing data were excluded. Data from large databases have been entered by multiple sources so threats to reliability and validity of data exist. Overtime and differential wage data were not obtained due to burden of extraction. The study was completed using data from a single unit in a one organization; hence the results of the study are not generalizable beyond the study unit. The methodology of using patient level data to explore direct nursing cost can be replicated and expanded using all units the patient is on during an acute care episode. Innovative patient assignment software provides a convenient source of data for nurse scientists and nurse leaders to use in creating next generation nursing science.

Keywords:
Patient level data; Research; Direct Nursing Cost
Repository Posting Date:
17-Nov-2014
Date of Publication:
17-Nov-2014 ; 17-Nov-2014
Other Identifiers:
INRC14J10
Conference Date:
2014
Conference Name:
25th International Nursing Research Congress
Conference Host:
Sigma Theta Tau International, the Honor Society of Nursing
Conference Location:
Hong Kong
Description:
International Nursing Research Congress, 2014 Theme: Engaging Colleagues: Improving Global Health Outcomes. Held at the Hong Kong Convention and Exhibition Centre, Wanchai, Hong Kong

Full metadata record

DC FieldValue Language
dc.language.isoen_USen_GB
dc.language.isoenen
dc.type.categoryFull-texten
dc.typePresentationen
dc.titleExploring Nursing Cost Using Patient Level Dataen
dc.title.alternativeExploring Technology to Improve Patient Careen
dc.contributor.authorJenkins, Peggy A.en
dc.contributor.departmentAlpha Kappaen
dc.author.detailsPeggy A. Jenkins, PhD, RN, pjenkins0701@gmail.comen
dc.identifier.urihttp://hdl.handle.net/10755/335055-
dc.description.abstractSession presented on Sunday, July 27, 2014: Purpose: A focus in the American healthcare system is on value-based delivery of services. Payment models will reward hospitals that can efficiently and effectively care for patients. Since nursing is the largest labor segment in the hospital, understanding this very important input is imperative in delivering high quality care at an affordable price. Traditional methods of accounting for nursing services have included measuring nursing hours per patient day, which is a unit level measurement that does not include variability in nursing care at the patient level. Nursing administrators, healthcare leaders, payers, and patients should be interested in understanding the cost of nursing service at the patient level. To benchmark patient level costs, hospitals need to move beyond current unit level measurements of cost to analytics linking individual patients to nurses. Using patient level data in nursing research provides a source capable of answering many unrequited questions about individual nursing contributions to patient outcomes. The current science of nursing cost will evolve and evidence can be provided through which value based healthcare is better built across the world. 1,2,3,4,5,6,7,8 The purpose of the study was 1) to explore the variability of nursing cost per acute care episode for patients with similar DRGs with and without major complications; 2) to investigate the relationship among patient characteristics, nurse characteristics, nursing intensity, and nursing cost as a patient outcome. Four research questions included: What is the variation in nursing cost per acute care episode for patients with the same DRGs without complications, with complications, with major complications? What is the relationship among nursing characteristics (age, years of service, educational degree) on nursing cost per acute care episode? What are the characteristics of nurses assigned to patients with complications and major complications? What is the average nursing cost per day measured at the patient level? Methods: A retrospective, exploratory, cost study using secondary patient level de-identified data was completed. The study site was one general medical surgical unit in a large academic medical center located in the Central United States. The study site organization has earned Magnet designation. Sample was 3111 patients and 150 nurses on the unit over a two-year period. Important and time-consuming steps in secondary data analysis included data acquisition and management. The collaborative data acquisition process was complex and took almost two years to accomplish. The source of data was Clairvia Care Value Management Patient Assignment software plus medical management and human resources databases. De-identification of data was completed before the researcher obtained data. Staff time to de-identify data was negotiated via conference calls and on-site meetings by the researcher with the president of the software company and the nurse scientist and other significant staff from the study site. The researcher used a model for data management consisting of four steps; 1) planning, 2) organization, 3) computing, 4) documentation. 9 Stata do-files provided a tool to systematically record commands for each step of the data management and analysis process. Forty-five variables were collected, cleansed, and new variables were constructed for a total of forty-nine final study variables. Data were analyzed at the shift level, day level, and patient episode of care level according to the research question. Shift level patient and nursing intensity data were one original source of data, which were aggregated to patient episode of care to understand nursing cost per acute episode. Patient and nurse characteristic data were another source of data that were merged with shift level data. The conceptual definition of nursing cost per acute care episode (NCACE) was comprised of lower level elements (NI-nursing intensity) that were aggregated to form the emergent model (summation of individual nurse intensity per patient multiplied by individual nurse hourly wage). NCACE represents Nursing Cost per Acute Care Episode, which was defined as the summation of the product of nursing intensity (NI) and nursing hourly wage (NW). <p align="center">NI= f (time spent with patient) + (skill level of nurse) Nursing intensity was a function of time spent with the patient measured using Clairvia demand-driven patient assignment software. Nursing intensity was calculated in the patient assignment software based on the following methodology. The acuity score was a 1-12 point scale derived from an outcomes-driven model grounded in the Pesut and Herman conceptual framework, the Outcome-Present State-Test Model of Clinical Reasoning. 10 Nurses rated patients every shift or when condition changed on several outcomes using a 1-5 point scale that contributes to an algorithm producing the nursing intensity score. A monthly audit was completed to assure interrater reliability of acuity measurement and the result was 86% accuracy. Skill level of the nurse was recorded in the Clairvia software as RN, LPN, and Patient Care Associate. NW= f (experience, years of service, education, certification) Nursing wage was operationalized using actual hourly wage for each nurse providing direct care. The principle diagnosis was measured through the DRG. Complications were measured using a four level variable 'compcode'. The four value labels of the variable were 0 = no drg, 1= without complications, 2 = with complications, 3 = with major complications. Using the same DRG without, with, and with major complications allowed for a comparison of nurses assigned to increasingly complex patients. Nurse characteristics were measured and analyzed in relationship to cost of nursing care. Variables describing the nursing unit that are contextual included unit type, number of beds, average acuity per patient, unit skill mix. Nurse characteristics measured included skill level, age, education level, years of service at institution, years of service on the unit, and float. De-identified data from three databases were merged into a single file and analyzed using Stata software. Correlation analysis and regression analysis were used to explore relationships among patient characteristics, nursing characteristics, and nursing cost per acute care episode. Microeconometric measurement was used to determine the elasticity of nursing characteristics on patient acuity and direct nursing cost per patient. IRB expedited approval and continuation was obtained from the study site and the researcher's University. Results: Key findings included 1) patients with the same diagnosis have large variability in nursing intensity and nursing cost by shift, day and acute care episode (i.e. cost per day range DRG 192 $5.68-287.37, 191 $5.96-257.56, 190 $10.06-366.86); 2) nurses may not be assigned patients based on experience and education level; 3) direct nursing cost per patient on the study unit was $96.48 on average per day, which was only 5.8-7.3% of the daily room and board charge. Conclusion: There is large variability in direct NCACE for patients with similar DRGs. An example is patients with COPD without complications (NCACE range $54-1570, M $325, SD $242); COPD with complications (NCACE range $17-3674, M 408, SD $427); COPD with major complications (NCACE range $132-1455, M $462, SD $316). Nurse scientists have provided evidence for variability in nursing cost for patients with similar DRGs for decades, yet hospitals in America continue to be reimbursed under an assumption that patients with similar DRGs receive the same amount of nursing care. 11,12,13,14 This study refutes the assumption. RN years experience in the organization was the nurse characteristic most associated with direct nursing cost. A 10% increase or 9.3 total nurse years experience in the organization for the patient episode of care is associated with a 9.9% or $34.92 increase in direct cost of nursing care per episode for patients on the study unit holding all other variables constant. Data did not support the hypothesis that nurses with greater experience or education level are assigned sicker patients. Average RN experience assigned per patient episode was not significant when regressed on average patient acuity. Percent of BSN nurses assigned was significant in the model with nominal effect. The mean nursing direct cost per day for all patients in the study was $96.48 ($55.73, range-$.33-600.81). The room and board charge for each patient in a medical/surgical unit at the study hospital ranges from $1321-1650 per day. Therefore, the direct nursing cost per day is only 5.8-7.3% of the daily room and board charge. Direct nursing care is a small percent of the cost, but patients don't know this because direct nursing cost is included in the room and board charge and not itemized on the patient bill. Limitations of the study include de-identified data from a secondary source were used and cases with missing data were excluded. Data from large databases have been entered by multiple sources so threats to reliability and validity of data exist. Overtime and differential wage data were not obtained due to burden of extraction. The study was completed using data from a single unit in a one organization; hence the results of the study are not generalizable beyond the study unit. The methodology of using patient level data to explore direct nursing cost can be replicated and expanded using all units the patient is on during an acute care episode. Innovative patient assignment software provides a convenient source of data for nurse scientists and nurse leaders to use in creating next generation nursing science.en
dc.subjectPatient level dataen
dc.subjectResearchen
dc.subjectDirect Nursing Costen
dc.date.available2014-11-17T13:43:04Z-
dc.date.issued2014-11-17-
dc.date.issued2014-11-17en
dc.date.accessioned2014-11-17T13:43:04Z-
dc.conference.date2014en
dc.conference.name25th International Nursing Research Congressen
dc.conference.hostSigma Theta Tau International, the Honor Society of Nursingen
dc.conference.locationHong Kongen
dc.descriptionInternational Nursing Research Congress, 2014 Theme: Engaging Colleagues: Improving Global Health Outcomes. Held at the Hong Kong Convention and Exhibition Centre, Wanchai, Hong Kongen
All Items in this repository are protected by copyright, with all rights reserved, unless otherwise indicated.