Newborn; Infection
Conditions
Keywords
NICU, POKE, Length of Stay, Infection, Neonates
Brief summary
To assess the impact of POKE on babies, the investigators will longitudinally track outcomes before and after implementation at Intermountain Healthcare's five NICUs. Process outcomes will include the number of total POKEs per baby and the number of painful POKEs per baby, each measured at both the patient-level and NICU-level. Clinical outcomes will include hospital acquired infections, length of stay, and mortality. Financial outcomes will include total variable costs and backfill rate. The effect of POKE on each of these outcomes will be measured using multivariable regression analysis with appropriate distributional families and interaction terms.
Detailed description
POKE was developed and implemented at Dixie Regional Medical Center's Neonatal Intensive Care Unit (NICU) over the past 10 years to eliminate waste and reduce harm in healthcare. POKE is a combination of a unique culture and process, with a supporting database, that is designed to guide and inform care decisions while minimizing POKEs. The program utilizes an implementation framework, educational materials, electronic health records (EHR), and decision support analytics. POKE's initial deployment showed extremely promising results for Intermountain, which included: (1) eliminating 11,000 POKEs per year (a 50% reduction in overall POKEs), (2) realizing $940,000 per year in cost savings (a 28% reduction of overall cost), (3) reducing length of stay by 2 weeks per average stay (a 21% reduction in length of stay), and (4) eliminating Hospital Acquired Infections (i.e., Central-line Associated Bloodstream Infection and Ventilator-associated Pneumonia), translating into 10 lives saved and a $5.2M savings over a decade. POKE will now be deployed and routinized within all Intermountain Healthcare NICUs and be developed as a commercial product for external customers. To assess the impact of POKE on babies, the investigators will longitudinally track several outcomes before and after implementation at Intermountain NICUs. Process outcomes will include the number of total POKEs per baby and the number of painful POKEs per baby, each measured at both the patient-level and NICU-level. Clinical outcomes will include hospital acquired infections, length of stay, and mortality. Financial outcomes will include total variable costs and backfill rate.
Interventions
PROCEDUREThere will not be an intervention, rather the investigators will deploy best practices and track POKEs within the healthcare system to evaluate clinical and operational outcomes.
There will not be an intervention, rather the investigators will deploy best practices and track POKEs within the healthcare system to evaluate clinical and operational outcomes.
Sponsors
Intermountain Health Care, Inc.
Study design
Observational model
COHORT
Time perspective
PROSPECTIVE
Eligibility
Sex/Gender
ALL
Age
No minimum to 12 Months
Healthy volunteers
Yes
Inclusion criteria
* All babies in Intermountain Healthcare NICU
Exclusion criteria
* None
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Number of POKEs per baby | 1 October 2018 - 31 May 2019 | The primary outcomes will be the number of total POKEs per baby and the number of painful POKEs per baby. We will assess the impact of POKE on each of these outcomes modeled as counts (ie, generalized Poisson distributions) using multivariable regression adjusting for potential confounders including age, gestational age, interaction terms, and nominal indicators of NICU (to account for baseline heterogeneity across sites). |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Total Variable Cost | 1 October 2018 - 31 May 2019 | Total variable costs (log-linear regression) |
| Aggregate Backfill Rate | 1 October 2018 - 31 May 2019 | aggregated backfill rate (quasibinomial regression) |
| Infection Rate | 1 October 2018 - 31 May 2019 | Also using multivariable regression, the secondary outcomes will be patient-level indicators, including hospital acquired infection (logistic regression) |
| Length of Stay | 1 October 2018 - 31 May 2019 | Length of stay (scaled beta regression) |
| Mortality | 1 October 2018 - 31 May 2019 | Mortality (logistic regression) |
Countries
United States
Contacts
Primary ContactR. Erick Ridout, MD
Backup ContactTerri Kane, RN, MBA
Outcome results
None listed