You are here

View Text Box

As the aggregator of healthcare information from patient electronic medical records and insurance claims for its region, HSX compiles COVID-related data daily in its clinical data repository. 

  • Through the strength of its Data Analytics and Quality program, the HSX team can customize reports across a variety of population-health and care-delivery dimensions of health that relate to combatting COVID-19 and its aftermath.
  • In its role as a convener for community-wide health issues, HSX brings a wealth of rich healthcare encounter data.
  • HSX supports surge planning for emergency department and inpatient visits; serves as a disseminator of data to researchers; and partners with a range of major entities involved in improving quality of care, community efforts, and operational efficiencies.

COVID-19 Trends

[note: these two views only, updated with data through 7/30/2020] 

Note fewer cases assigned to weekend dates, partly due to lab tests resulting predominantly on weekdays.  Also, while the number of new cases may seem to decline on some most-recent days, HSX experiences a lag in identifying COVID patients until we receive coded diagnoses at or shortly after discharge. Since the peak of April 27, the number of new cases of COVID diagnosed patients per day, across a seven-day average, had been dropping into July but may now be on the increase again.  In the week ending 7/30/20, HSX data showed 501 new cases reported (compared to 395 for the week ending 7/22), at an average of 73 new cases per day (compared to 56 for the week ending 7/22).



Each day, since March 19, more COVID-19 cases have been identified in the Delaware Valley, contributing to the cummulative total.   The rate of increase of this total, though, slowed in April and into May and appears to have flattened from June into July.   
Left chart, shows cumulative number of unique patients.  Right chart is the LOG 10 transformation of the cumulative number of unique patients by day, showing the relative rate of change over time.  The X axis is the exponent of the rate of change.  In other words, this chart shows cases adjusted for exponential growth to make it easier to see the overall trend in acceleration or de-acceleration of the outbreak.  A flatter curve means the number of new cases is slowing.  A flat curve means the number of new cases is steady, and a dropping curve means that the number of new cases has declined.

HSX is adding new data feeds reflected in the data.  Diagnoses in HealthShare data also increasingly reflect incorporation of laboratory data reaching HSX through hospital EMRs.  HSX will be receiving more data from lab testing companies, which will increase visibility of C19-positive patients not requiring hospitalization.
Note: cumulative total for cases here differs from those reported by the state and county health departments, which receive more complete lab-testing and outpatient data.  Facilities without credible volumes of coronavirus codes are also not included in the above data.

COVID-19 Population Metrics


Daily in patient admissions and discharge

Red line represents admissions; green line discharges.  Emergency department (ED) volume was higher in May, presumably and partly reflecting more routine (normal, non-COVID-19) visits resuming.  ED volume on 5/18 was the highest over the past several weeks, with more than half the hospitals showing above average volumes based on the prior 30-day average. Inpatient admissions are continuing a gradual but steady climb, week to week, as hospitals resume normal operations; however, they remain well below pre-outbreak volume.  Inpatient admissions on May 18 were at the highest level in the past 8 weeks.

Daily cumulative increase/decrease in hospital inpatient bed census

Inpatient census steadily decreased since the height of the pandemic in mid-April, as hospitals continued to delay many normal services, including elective surgery and other procedures requiring hospital stays, and discharging patients who had previously received these services –– but with that trend showing signs of beginning a reversal in mid-May.


The mode (most common) length of stay (LOS) for inpatient COVID-19 treatment remains two days; however, LOS shows significant variation, with many lengths of stays longer, resulting in a median LOS of five days across all ages and races.  Older patients tended to have longer stays.  (Deceased patients are excluded from these data.)


[Note: Reports from HealthShare Exchange’s (HSX) Data Analytics and Quality department are based on trends and activity evident in data contributed by HSX provider members across the region, including parts of New Jersey, Pennsylvania, and Delaware.  Data shown come from clinical encounters (i.e., ADT – Admit, Discharge, Transfer HL7 messaging) and lab results sent to HSX directly from members’ electronic medical records (EMR) systems or from testing labs.  Results graphed on this page are through May 19, 2020, unless otherwise noted.​

COVID-19 diagnoses based on ––
• COVID-19 ICD-10 code on an outpatient, ED or inpatient encounter: B97.29, B34.2, U07.1 and U07.2 (the U codes are the CDC recommended codes starting April 1, 2020).
• Positive lab results for any version of the SARS-CoV-2 test.

These data totals above, aggregated for HSX-member healthcare entities, may vary from governmental or other organizations’ data sets, as HSX  information is collected through health information exchange and represents activity among participating facilities based on ADT feeds and thus does not fully account for ambulatory patient activity within HSX’s medical service area.]