HASA LACE Preliminary Readmissions Prediction Documentation & Validation

HASA introduced the LACE report in its reporting suite in January, 2017.  This report complements HASA’s capability to support organizations to look back at readmissions across organizations and identify potential readmits before the readmission occurs. The LACE report identifies patients’ likelihood of readmission or death in the 30 days following a hospital discharge based on Length of Stay (LOS), Acuity, Co-morbidities, and previous Emergency visits. Researchers1 developed this 19 point scale, with 0 having a 2% likelihood and 19 having a 43% likelihood of readmission or death in 30 days following the discharge (Figure I).

Figure I.

Based on data representing patients in the derivation and internal validation groups. Note: bars = number of patients with the same LACE score; black line = expected risk of death or unplanned readmission within 30 days after discharge; grey line = observed risk (error bars = 95% confidence intervals).


HASA Validation

With HASA’s ability to identify the four LACE criteria, a validation study was conducted using the following methodology:

  • HASA complied patients from all 35 participating San Antonio hospitals with discharge dates of June 1 and July 1, 2016.
  • LACE scores were calculated for each discharged patient based on the LOS, acuity (whether they were admitted via ED), number of comorbidities, and number of previous ED visits in the previous 6 months.
  • Actual readmissions for each patient to same or other hospital were calculated for each patient and displayed as a percentage, and averaged for each LACE score.
  • A trend line was applied for comparison to the original findings (Figure II).

Figure II.

  • HASA results indicated a comparable trend of increased patient readmissions with a higher calculated LACE score. HASA’s trended maximum score was slightly lower at the high end of the scale (34% vs. 43%) than published. This may be attributed to the fact that HASA can only validate readmissions and not death rate, as in the original study.


Social Determinants for Readmissions

Given that HASA aggregates patient information from across the community, the validation process was extended to assess the potential impact on readmissions rate as determined by the LACE score for patients living in highly vulnerable vs. moderate and low vulnerable zip codes (based on income and education level) of the community. 

  • Discharged patients for four successive dates (March 1, April 1, May 1, and June 1, 2016) from all 35 participating San Antonio hospitals (N=3200) were categorized by zip codes, submitted to the LACE algorithm, and scored on readmissions likelihood. The zip codes were then stratified according to a high, moderate, and low vulnerability categories2.
  • Figure III, below, offers preliminary indications that correlative trends between readmissions and LACE score are consistent and more pronounced in the highly vulnerable zip code region. This finding can aid discharge planners and care teams in focusing intensely on the high risk LACE scoring cohort residing in highly vulnerable regions of the community.

Figure III.


1Van Walraven, Dhalla, et al. “Derivation and Validation of an Index to Predict Death or Unplanned Readmissions after Discharge from Hospital to Community.” Canadian Medical Association Journal. April, 2010. 

2American Community Survey (ACS) - 2010-2014 five-year estimates at the ZIP Code Tabulation Area (ZCTA) level