Human capital in the nursing workforce and its impact on patient outcomes
This presentation is the property of its rightful owner.
Sponsored Links
1 / 27

Human Capital in the Nursing Workforce and Its Impact on Patient Outcomes PowerPoint PPT Presentation


  • 49 Views
  • Uploaded on
  • Presentation posted in: General

Human Capital in the Nursing Workforce and Its Impact on Patient Outcomes. Ciaran S. Phibbs, Ph.D. June 2, 2007. Research Team. Investigators Ann Bartel Patricia Stone Nancy Beaulieu Programmers and Research Assistants Lakshmi Ananath Cecilia Machado Susan Schmitt Andrea Shane.

Download Presentation

Human Capital in the Nursing Workforce and Its Impact on Patient Outcomes

An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -

Presentation Transcript


Human capital in the nursing workforce and its impact on patient outcomes

Human Capital in the Nursing Workforce and Its Impact on Patient Outcomes

Ciaran S. Phibbs, Ph.D.

June 2, 2007


Research team

Research Team

  • Investigators

  • Ann Bartel

  • Patricia Stone

  • Nancy Beaulieu

  • Programmers and Research Assistants

  • Lakshmi Ananath

  • Cecilia Machado

  • Susan Schmitt

  • Andrea Shane

Health Economics Resource Center


Background

Background

  • Relationship between hospital nurse staffing and quality of care is a significant concern for health services researchers, health services providers and policymakers.

  • Prior research has found evidence showing negative correlation between adverse patient outcomes and nurse staffing levels, but this literature has numerous limitations.

Health Economics Resource Center


What our paper contributes

What Our Paper Contributes

  • Unit-level data

  • Monthly observations

  • Longitudinal 4 years.

  • Data from a single organization (VA).

  • Adds role played by human capital and relational capital in the nursing workforce

Health Economics Resource Center


Hypotheses we test

Hypotheses We Test

  • RNs with more general human capital (education, experience) deliver higher quality nursing care

  • RNs with more firm-specific human capital (job tenure) deliver higher quality nursing care

  • Relational capital (stability of nursing team) contributes to higher quality care because stable teams are better at sharing tacit knowledge and facilitating coordination.

Health Economics Resource Center


Nurse staffing data

Nurse Staffing Data

  • DSS, VA’s implementation of TSI, a comprehensive hospital activity based accounting system.

  • Inpatient nursing labor tracked at the unit level.

Health Economics Resource Center


Nurse staffing data1

Nurse Staffing Data

  • Monthly extract that reports the number of hours, for each type of nursing labor, including:

    • RNs

    • LVNs

    • Aides

    • Advanced practice nurses

    • Others

Health Economics Resource Center


Nurse staffing data2

Nurse Staffing Data

  • Hours worked and costs reported separately for “regular” hours, overtime, and vacation/sick/holiday hours. Thus, we can track nursing labor by the actual hours worked, not the hours nurses were paid.

Health Economics Resource Center


Human capital data

Human Capital Data

  • We extracted human capital and other nursing characteristics from the VA payroll data (PAID). These data contain a wealth of information, including VA tenure, education, and age.

  • PAID also reports hours of overtime and shift premiums (track off-shift labor).

Health Economics Resource Center


Human capital data1

Human Capital Data

  • DSS tracks where people work to allocate costs.

  • DSS made a special extract for us that summarized, by PAID codes, how all nursing labor was allocated across units, for each facility.

Health Economics Resource Center


Patient data

Patient Data

  • Unlike most discharge abstracts, VA has a “bedsection” file, with one record for each bedsection stay, which are defined by the treating specialty of the physician. ICUs are separate bedsections.

    • Thus, most, but not all, unit stays have a corresponding bedsection record.

Health Economics Resource Center


Empirical model

Empirical Model

  • PSIit = α1(RN Hrs PPD)it + α2(RN Hrs PPD2 )it +

  • α3 HCit + α4 Non-RN Hours PPDit + α5DRGit + α6AGEit + α7 LOSit + α8Dischargesit +

  • Year + λi + εit

  • HC is measured by percentage of RN hours provided by RNs with at least a B.S. degree, RN age (“experience”), RN tenure, and percentage of RN hours provided by part-time RNs.

Health Economics Resource Center


Summary statistics

Summary Statistics

Health Economics Resource Center


2006 average rate of failure to rescue by icu bed section

2006 average rate of failure to rescue, by ICU bed section

Health Economics Resource Center


2006 average rate of infection by icu bed section

2006 average rate of infection, by ICU bed section

Health Economics Resource Center


Preliminary findings

Preliminary Findings

  • General human capital (education) is negatively correlated with the rate of infections due to medical care

  • Specific human capital (job tenure) is negatively correlated with the rate of infections due to medical care and rate of failure to rescue. The former correlation holds even controlling for nursing-unit fixed effects.

  • Data are only for intensive-care units and results are likely to be a lower bound on effects for more diverse units.

Health Economics Resource Center


Summary of ols results human capital variables

Summary of OLS Results: Human Capital Variables

  • General human capital (education) is negatively correlated with the rate of infections

  • Specific human capital (tenure) is negatively correlated with the rate of infections. A one standard deviation increase in RN tenure is associated with a 19% decrease in rate of infections

  • Specific human capital (tenure) and failure to rescue are negatively correlated (though less robust). A one standard deviation increase in RN tenure is associated with a 4% decrease in failure to rescue rate.

Health Economics Resource Center


Summary of ols results other variables

Summary of OLS Results: Other Variables

  • RN staffing has a negative but diminishing effect on infection rate, and is negatively correlated with failure to rescue rate

  • Non-RN hours has no effect

  • Very low patient volume is associated with higher failure to rescue rate

Health Economics Resource Center


Summary of fixed effects results

Summary of Fixed Effects Results

  • RN tenure negatively correlated with rate of infections and the measured impact is more than twice as large as the OLS effect.

  • RN tenure is insignificant in failure to rescue equation.

  • RN education is insignificant in both equations.

Health Economics Resource Center


Plans for future work

Plans for Future Work

  • Expand sample to include all acute-care units

  • Control for prior experience of RNs

  • Control for union membership

  • Study other measures of health care quality such as 30-day mortality rate, re-admission rates and patient-assessed quality of care

  • Add more information about human capital, especially for off-shifts

  • Add information from RN satisfaction survey

  • Construct measures of team stability to study relational capital

Health Economics Resource Center


Human capital in the nursing workforce and its impact on patient outcomes

Health Economics Resource Center


Patient data1

Patient Data

  • 88% of the acute bedsection records link to only 1 IPD stay.

    • 437,465 acute hospitalizations in FY 06

    • 54,323 had more than 1 IPD stay mapped to the same bedsection stay

      • 94% of these had 2 IPDs

      • 5.7% had 3 IPDs

      • 0.3% had 4 or 5 IPDs

Health Economics Resource Center


Correlations

Correlations

Health Economics Resource Center


Rate of infections ols regressions

Rate of Infections, OLS Regressions

Notes: N=2866 in columns (1) through (4). N=2821 in column 5. All regressions include dummy variables

for quartiles of the number of discharges and year dummy variables.

Health Economics Resource Center


Failure to rescue rate ols regressions

Failure to Rescue Rate, OLS Regressions

Notes: N=2673 in columns (1) through (4); N=2631 in column (5). All regressions include dummy variables for quartiles of discharges and dummy variables for years.

Health Economics Resource Center


Rate of infections fixed effects

Rate of Infections Fixed Effects

Notes: N=2866 in columns (1) through (4). N=2821 in column (5). All regressions include dummy variables for quartiles of discharges and dummy variables for years.l

Health Economics Resource Center


Failure to rescue rate fixed effects

Failure to Rescue Rate, Fixed Effects

Notes: N=2673 in columns (1) through (4). N=2631 in column (5). All regressions include dummy variables for quartiles of discharges and dummy variables for years.

Health Economics Resource Center


  • Login