2000 Michigan Hospital Report                                        April 2000
Questions and Answers
          

 

Technical Description and Methodology for Public Data Reports

             The methodology for this report was developed by the MHA Data Issues Committee and its Technical Advisory Group.  The data cover the timeframe of the calendar year of 1998.  The following is a detailed description of the process for inclusion in each of the indicator measures.

How severity-adjustment works

             The 3M All Patient Refined Diagnosis Related Group (APR-DRG) system classifies each record into one of four subclasses for a given APR-DRG.  The next step is to calculate Michigan-specific norms for Length of Stay and Mortality by APR-DRG subclass.  These norms are used in an algorithm that calculates the expected ranges for the measures that combine several APR-DRGs.  The expected range, a 95 percent tolerance interval, narrows as the volume of cases increases.

 Methodology

 

Global Exclusions

 ·Cases where the Length of Stay was zero or greater than a year. 

·Patients discharged from rehabilitation facilities.

·Outliers for Length of Stay (identified by being outside the 95 percent confidence interval of the mean for each APR-DRG subclass).  The Length of Stay outliers are not excluded from the mortality measures.

·Replicated cases.  (When a hospital cannot submit data on all of its cases for some reason, the portion of the data that has been submitted is replicated to simulate a full year’s representation of volume in the Michigan Inpatient Data Base.  Because these are not actual cases, they are not included in the analysis.  Less than 1 percent of cases in the MIDB are replicated.)

 General Indicators

 Selected Medical Cases

            The case must be classified as one of the following APR-DRGs: 045, 046, 138, 139, 140, 141, 144, 241, 242, and 244.  In general, these translate into the following HCFA DRGs: 014, 015, 089, 090, 091, 096, 097, 098, 099, 100, 101, 102, 174, 175. 

 Selected Surgical Cases

            The cases must be classified into one of the following: A) APR-DRGs 023, 120, 121, 160, 167, 168, 169, 173, 174, 175, 221, 223, 303, 304, 310, 446; B) principal procedure of 32.29 or 32.3x with an APR-DRG of 121 or 122; or C) principal procedure of 68.3x, 68.4x, 68.5x, 68.6x, or 68.7x with an APR-DRG of 511, 512, or 513.  The cases in Group A generally fall into the HCFA DRGs of 004, 108, 110, 111, 112, 148, 149, 152, 153, 214, 215, 310, or 311.

 Obstetrics

 Cesarean Sections (C-sections)

            This is calculated by taking all C-sections performed and dividing by the number of deliveries.  The numerator includes cases with DRG 370 or 371.  The denominator includes cases in DRGs 370-375.  No severity adjustment was performed.

 

Vaginal Birth After Cesarean (VBACs)

            VBACs are calculated by taking all VBACs performed and dividing by the number of women with previous C-Sections.  The numerator includes those cases in DRGs 372-375 and a diagnosis code starting with 654.2 (the principal diagnosis and up to 14 secondary diagnosis codes were examined).  The denominator includes those cases with DRGs 370-375 and a diagnosis code starting with 654.2.  No severity adjustment was performed.

 Cardiac Care

 Nonsurgical Heart Care

            These cases must fall into one of the following APR-DRGs: 190, 191, 192, 193, 194, 198, 199, 200, 201, 202, or 203.  These roughly correspond with the following HCFA DRGs: 121, 122, 123, 124, 125, 126, 127, 132, 133, 135, 136, 137, 138, 139, 140, and 143.

 Coronary Artery Bypass Grafts (CABGs)

            These cases must fall into one of the following APR-DRGs: 165, 166.  These are similar to HCFA DRGs 106 and 107 — some 108 and 109 cases may be included.

 Valve Repair

            These cases must fall into one of the following APR-DRGs: 162, 163.  These are similar to HCFA DRGs 104 and 105 — minus cases with cardiac defibrillator procedures.  These cases, though HCFA includes them in the valve repair DRGs, are not true valve procedures.  The APR system categorizes these cases into APR-DRG 161, which is not included as part of the definition of this indicator.

 Joint Replacement

 Total Hip Replacement

            In order to be included in this measure, a case must meet the following criteria:  principal procedure of 81.51; not a principal diagnosis beginning with 820, 821, 822, or 823; and discharged alive.  The data were then severity adjusted and all but a few fell into one APR-DRG (212); those few that did not fall into that APR-DRG were excluded.

 Total Knee Replacement

            In order to be included in this measure, a case must meet the following criteria:  principal procedure of 81.54; not a principal diagnosis beginning with 820, 821, 822, or 823; and discharged alive.  The data were then severity adjusted and all but a few fell into one APR-DRG (212); those few that did not fall into that APR-DRG were excluded.

 

 

 

[ Home | Table of Contents | Order | MHA

Michigan Health & Hospital Association