Overview
Trella Health analyzes billions of claims each quarter to produce the data in Marketscape. This data is updated quarterly about 4 months after the end of the calendar quarter being reported. See Data Release Timeline. There are some common parameters that we use throughout Marketscape for all calculations.
Medicare claims have three date related metrics.
- Claim "From" date - which is the start of service.
- Claim "Through" date - which is the end of service.
- Claim "Processing" date - which is tied to went claim was processed.
A "stay" or "period of service" is from admission to discharge; the "from" date to the "through" date. An outpatient stay, or a visit to a doctor's office will have the same "from" and "through" date. So, if you imagine a count of patients treated by a physician, we will count each patient treatment using the "from" date. And admission to hospice will use the "from" date. A hospital discharge will use the "through" date. And an Average Length of Stay will require both dates.
Although this may seem to be a bit "mathy," understanding the concepts below will help you to understand our metrics, how they are calculated, and how to derive the most value from them. Specifically, this article will help you to understand the definitions that will pop up in Marketscape when you click on the "i" in the column headers.
These are some of the standard parameters we use to define the claim statistics:
- There are events that trigger relevance which we include in a calculation. Some examples would be a patient's death or an admission to post acute care.
- There are reporting periods that sets boundaries for including specific events in counts and calculations. There are three reporting periods that are described in more detail below.
- There are some requirements that limit each statistic. A common example is 3 Months Prior, which requires a physician claim within a 3 month reference period to the post acute care admission event.
We have included this information so that you can understand the definitions included in Marketscape and so that you can accurately compare the different statistics while keeping the time periods clear.
Events
There are numerous data events that become a reference point for determining whether specific claim data is included in Marketscape and also how it is used. Some examples are:
- A patient's death suggests the level of medical acuity and also indicates a high appropriateness for hospice care. In mortality counts we track physicians who treated patients within the six month reference period prior to the date of death. See Critical Concept - Mortalities.
- Discharge from inpatient or post acute care. The date that a patient was discharged from inpatient care is used for determining the significance of other post-discharge events. The most common usage is the readmit rate which calculates the percentage of a facility's patients who were discharged from a hospital, were admitted to post-acute care and then re-admitted to a hospital during the 30 day post discharge reference period. In that example, the date of discharge from post acute care is also a required event. Hospitalization rates are also calculated based on a 30 day reference period from the date of discharge from post-acute care.
- Re-admission to Inpatient Care. Any time a patient is re-admitted to inpatient care within 30 days from the initial discharge this is counted against the hospital. For this to count as a re-admission, the discharge and re-admission need to both fall in the same 30 day reference period.
- Admission to post acute care. Since the data in Marketscape is in reference to medical care provided by Hospices and Home Health Agencies, the date of admission to post-acute care is critical. Based on these admissions, the informal patient/physician referral relationship can be established by counting physician claims within the three month reference period prior to PAC admission. This also allows us to identify the last submitted claim prior to PAC admission.
- Discharge Status and Primary Diagnosis - although these are not "events," for many claims we track status codes or diagnoses to differentiate populations of patients in many tables.
Reporting Periods
There are two reporting periods that are used throughout Marketscape. A reporting period is limited by a specific start and end date. Only claims dated within the reporting period will be used for compiling data. For our understanding in this article, these are reference claims because we often analyze other claims with reference to them.
There are several standard time periods that we use in calculating the metrics throughout the Solutions. This article lists the most important time periods used for calculating various metrics.
One-year reporting period:
Most of the counts that are reflected by whole numbers are based on claims data compiled from a one-year reporting period. The one-year reporting period often ends on the last day of the chosen quarter and includes the period combining the chosen quarter and the previous three quarters. See Graphic:
For example: if the latest data release ends with 2019-Q2, the data throughout the Solution would reflect claims from the following four quarters: 2019-Q2, 2019-Q1, 2018-Q4, and 2018-Q3: From July 1, 2018 through June 30, 2019, inclusive.
Two Year Reporting Period
Most performance metrics that are presented as percentages (%) in Marketscape are calculated on a two-year reporting period. The two-year reporting period ends on the last day of the most recent quarter of data and includes the period combining the chosen quarter and the previous 7 quarters. This expanded reporting period is used to incorporate the larger sample size that is required to produce valid metrics.
For example: if the latest data release ends with 2019-Q2, the data throughout the Solution would reflect claims from the following four quarters: 2019-Q2, 2019-Q1, 2018-Q4, 2018-Q3, 2018-Q2, 2018-Q1, 2017-Q4, and 2017-Q3: From July 1, 2017 through June 30, 2019, inclusive.
We are required to use a different reporting period for performance metrics in many cases to maintain privacy suppression. Imagine that we told you a doctor treated 67 patients and that 15% of those patients were readmitted. Math tells us that 15 percent of 67 is ten patients, which violates privacy suppression because we are identifying a number less than eleven. If the percent is calculated from a 2 year period, there is no way of knowing how many patients would be included in the population used for the initial calculation.
Reference Periods
In some cases, a claim, or a specific data point on a claim is measured against another claim or event that represents a significant related care event. Sometimes the relevance of this connection is based upon the length of time between care events, for example, discharge from Inpatient care with reference to admission into PAC care. In other cases, a “reference” claim could be evaluated relative to another event that falls outside of the reporting period. In all of these circumstances, the column definition should identify the reference period.
Examples:
- 3 months prior - When calculating this data point, we start with an event - admission to post acute care - and we set a 90 day reference period prior to the admission. Only claims submitted during this reference period are included in the 3 month prior count.
- Mortalities - When calculating this data point, we start with an event - a patient death - and we set a 6 month reference period prior to death. Only claims submitted during this reference period are included in the mortalities count.