1999 Annual Report
THE SPARCS DATA SYSTEM
The Statewide Planning and Research Cooperative System (SPARCS) was implemented by the New York State Department of Health in 1979, with the cooperation and initial financial support of the U.S. Department of Health and Human Services. SPARCS receives, processes, stores, and analyzes the following: inpatient hospitalization data from all Article 28 facilities in New York State and ambulatory surgery data from hospital-based ambulatory surgery services and all other facilities providing ambulatory surgery services.
SPARCS continues to be a comprehensive, integrated information system available to assist hospitals and organizations in the health care industry with health care resource planning, financial analysis, decision making, and surveillance of New York State hospital and ambulatory surgery services and costs. SPARCS has proven to be an effective management tool, not only for the Department of Health but also for the health care industry. Widespread support and advice from many organizations and individuals in the public and private sectors have made possible the development and refinement of SPARCS. The Department of Health continues to invite active participation in improving the quality and usefulness of SPARCS.
The 1999 Annual Report represents the twentieth full year of SPARCS data collection efforts. The Annual Report Series presents hospital inpatient stay data based on discharges for each year through a set of standard statistical tables which serve the needs of a wide spectrum of health information users.
From 1980 through 1993 SPARCS made use of two data sources: the Discharge Data Abstract (DDA) and the Uniform Billing Form (UBF). In 1994 SPARCS began collecting essentially the same information from a single source based on the Universal Data Set (UDS) specifications. These specifications blend the UB-92 nationwide inpatient and outpatient billing requirements with the unique billing and discharge data reporting requirements of New York State. The single UDS data stream requires that medical abstract information and billing data are merged before they are sent to SPARCS.
This electronic format streamlines multiple data submission formats into a single format, removing redundant reporting requirements for hospitals and other health care facilities, while continuing to support the myriad of requests from health care researchers for both billing and medical records data. Appendix A lists the UDS data elements collected by SPARCS.
Each health care provider submits its SPARCS data in the uniform, computer-readable format described in the UDS. The data are sent to the Department of Health either directly by the hospital or through one of a number of private information processing services. Every record received is edited to identify errors, and hospitals are notified of records needing correction. Each data element must have a valid value before the record is accepted by the system. When a record needs correction, the hospital or processing service is notified. Duplicate submissions are carefully screened.
Regulations governing the confidentiality of SPARCS data were adopted by the New York State Hospital Review and Planning Council with the advice of all sectors of the health care industry. The regulatory, tracking, and monitoring functions of SPARCS are administered by the New York State Department of Health. The responsibility for protecting the confidentiality and privacy of data related to patient care resides with the Commissioner of Health.
To protect patient privacy, patient names are omitted from the SPARCS data set. The focus of the system is the incidence of diseases or conditions requiring hospitalization rather than individual patients. For this reason, users of SPARCS data cannot ascertain the number of individuals treated for a specific disease, only the number of hospitalizations that have occurred.
USING THE 1999 SPARCS ANNUAL REPORT
Descriptions of the data displayed in the tables are included below to assist in using the report:
Expected Primary Source of Reimbursement
The expected primary source of reimbursement used in the 1999 Annual Report is obtained from the Universal Data Set. This data element is documented at the time of discharge, not the time of payment, and represents the best information available to the hospital when a patient leaves the facility. However, given the complexity of reimbursement processes, especially when a patient appears to be eligible under more than one third-party payment plan, the expected primary payer is not always the ultimate primary payer. There may also be a bias toward under reporting of discharges for third-party payers when there are delays in establishing eligibility.
Special care should be taken in interpreting tabulations of patient days by expected primary source of reimbursement. For each discharge a single payer is reported as the expected primary source of reimbursement. Since many cases have multiple payers, the expected primary payer may not pay the entire bill. This information is displayed on Tables 1, 9, and 15.
The service categories used in the 1999 Annual Report are based on categories developed by the New York State Department of Health and are defined in terms of diagnosis and procedure codes from the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM). This information is displayed on Tables 1, 2, 3, 5, 8, 11, and 12.
NURSERY - Patients whose age is equal to zero and have an ICD-9-CM code of V30.0, V30.1, V31.0, V31.1, V32.0, V32.1, V33.0, V33.1, V34.0, V34.1, V35.0, V35.1, V36.0, V36.1, V37.0, or V37.1 are considered nursery patients. These codes refer only to live-born infants. Definitions of nursery service category and newborn age category are the same.
OBSTETRICAL - Obstetrical patients are females of any age with an ICD-9-CM code within the range of 630 through 634.9 and 640 through 676.9.
PSYCHIATRIC - Psychiatric patients are any age with an ICD-9-CM code within the range of 290 through 319.
PEDIATRIC - Pediatric patients are age 14 years and younger whose primary diagnosis is neither included in the nursery, obstetrical, or psychiatric categories nor classified by ICD-9-CM codes 614 through 629.9 (gynecological) or 635 through 639.9 (abortion).
MEDICAL - Medical patients are age 15 years or older with no reported procedure code or with reported procedure codes greater than 86.99. This category also includes gynecological patients (ICD-9-CM codes 614 through 629.9) of all ages with no reported procedure code or with reported procedure codes greater than 86.99. Excluded are newborn, obstetrical, psychiatric, and pediatric patients as defined above.
SURGICAL - Surgical patients are age 15 years or older with at least one procedure code in the range 01 through 86.99. This category also includes gynecological patients (ICD-9-CM codes 614 through 629.9) of all ages with at least one such procedure code and all patients whose principal diagnoses are codes 635 through 639.9 (abortion). Excluded are newborn, obstetrical, psychiatric, pediatric, and medical patients as defined above.
Disposition of Patient
The 1999 Annual Report includes disposition of patient information which is obtained from the Universal Data Set. It identifies the patient's destination or status upon discharge. The information is displayed on Tables 1 and 16. The categories used in the report include home, another acute care hospital, skilled nursing facility, intermediate care facility, other institution, home health services, left against medical advice, psychiatric chronic care, and died. To reduce the number of categories displayed in the tables, some are grouped together. "Neonate Discharged to Another Hospital for Neonatal Aftercare" and "Transferred to Another Hospital for Tertiary Aftercare" are included in "Another Acute Care Hospital". "Discharged Under Care of Home IV Provider" is included in "Home Health Services". "Admitted to Domiciliary Care Facility" is included in "Other Institution". "Discharged to Intermediate Care Facility for the Mentally Retarded" is included in "Intermediate Care Facility".
Length of Stay Calculation
The 1999 Annual Report includes length of stay information. SPARCS calculates a length of stay for each discharge record by subtracting the date of admission from the date of discharge. If a patient is admitted and discharged the same day, the length of stay is one day. This information is displayed on Tables 1 - 9, 11, and 14 - 17.
Diagnosis Related Groups
Tables 14, 15, and 16 display information on Major Diagnosis Categories (MDCs) and Diagnosis Related Groups (DRGs). DRGs are a classification system used to categorize patient discharge abstracts into meaningful groupings. Federal MDCs and DRGs are displayed in the tables. New York State MDCs and DRGs are available upon request.
The 1980 to 1982 Annual Reports used the "original" 383 category DRG model, which was developed before the implementation of ICD-9-CM coding and depends on the conversion of codes to earlier coding schemes. The 1983 to 1998 reports used versions of the "new" DRG model which was designed to use ICD-9-CM codes directly. The first "new" DRG model had 470 DRGs arrayed within 23 Major Diagnostic Categories (MDCs). This version was used for the 1983 to 1985 reports. Because of new technologies, new coding for diseases, and the use of DRGs for reimbursement, adjustments have been made to the grouping methodology in new versions. The 1986 report used the 471 category revision, the 1987 report used the 473 category revision, and the 1988 report used the 475 category revision. In 1989 and 1990 the number of categories was 477 for both years though they were separate revisions. The 1991 report used the 490 category revision, which involved a major restructuring of the overall DRG classification scheme. The 1992 and 1993 reports used different revisions of the 492 categories which incorporate the major changes introduced in the previous revision. The 1994 report used the 494 category revision, and the 1995, 1996, and 1997 reports all had 495 categories, though they were separate revisions. In the 1998 report, the number of DRG categories increased to 503, with the addition of five DRG categories (496-500) dealing with spinal fusion and three DRG categories (501-503) dealing with knee procedures. For the 1999 report, the number of DRG categories expanded to 511, with the addition of eight DRG categories (504-511) dealing with burn diagnoses. For 1999 the number of DRG categories remained at 511.
The earlier DRG revisions began the classification by categorizing all principal diagnoses into 23 mutually exclusive and exhaustive MDCs, based on predetermined riteria. Within each MDC, the criteria used to select the DRG for a record includes the principal diagnosis, secondary diagnosis, operating room procedures, the presence or absence of a substantial comorbidity and/or complication, age, and discharge status. Invalid or clinically inconsistent information is classified in either DRG 468 (Extensive Operating Room Procedure Unrelated to Principal Diagnosis), DRG 469 (Principal Diagnosis Invalid as Discharge Diagnosis), DRG 470 (Ungroupable - Discharge with Invalid Data), DRG 476 (Prostatic Operating Room Procedure Unrelated to Principal Diagnosis), or DRG 477 (Non-Extensive Operating Room Procedure Unrelated to Principal Diagnosis).
DRGs 103, 468, 469, 476, 477, 480, 481, 482, 483 and 495 are not associated with a specific MDC and each case is reported in the appropriate MDC based on their diagnoses, procedures, and other criteria. DRG 470 is reported in MDC 00.
Figure A illustrates the method of selecting one of the eight DRGs specific to MDC 16: Diseases and Disorders of the Blood and Blood-Forming Organs and Immunity.
Average Total Charge of Stay
The 1994 Annual Report is the first report to display charge information. Charge information is displayed with MDC, DRG, diagnostic, and surgical procedure data on Tables 13 and 18. Charges include both the covered and non-covered portions of patient stay for ancillary services and accommodations. Covered charges are those charges reimbursable by the primary payer.
COUNTY OF RESIDENCE
The 1999 Annual Report displays information on patient's county of residence on Tables 7, 8, 11, and 12. There are 128,267 discharges coded as having an unknown county of residence. Over fifty percent of these discharges are from four hospitals, who coded all their discharges as having an unknown county of residence. Detailed information on this data is available by contacting the SPARCS Administrative Unit.
SPARCS DATA REQUESTS
SPARCS information is available on the Department of Health Web site:
Annual Reports for 1994 - 1999 are viewable on this Web site. Reports for 1994 - 1999 are available for downloading in WordPerfect format from this site also. In addition, tables (beginning with 1987) are also available in spreadsheet format.
Any questions regarding SPARCS data collection should be directed to:
|SPARCS Administrative Unit
Bureau of Production Systems Management
New York State Department of Health
Empire State Plaza
Albany NY 12237-0023
|Phone: (518) 473-8144
Fax: (518) 474-9168
Any specialized requests for SPARCS data should be directed to:
|Bureau of Biometrics and Health Statistics
New York State Department of Health
Empire State Plaza
Albany NY 12237-0044
|Phone: (518) 474-3189
Fax: (518) 486-1630
When specialized requests are approved for production, a cost estimate is provided based on the availability of existing reports and the cost of analysis, programming, and computer time.
1999 HOSPITAL INPATIENT DISCHARGES
Reporting Data Completeness Issues
The Department of Health Data Quality Coordination and SPARCS Units analyze SPARCS reporting. As of 11/15/2000, the 1999 file was estimated to be approximately 99.2% complete. This file was used to create the Annual Report Tables.
The following hospitals have one or more months identified with a probable under reporting of discharges and have been contacted by SPARCS:
|Arden Hill Hospital||No data submitted for July.|
Low discharge for October.
|Community Memorial Hospital||No data submitted for November and December|
|Critical Access Hospital||Low discharges for November and December.|
|E. J. Noble Hospital Gouverneur||Low discharges for December.|
|Eastern Long Island Hospital||No data submitted for October.|
Low discharges for August, September, November and December.
|Medina Memorial Hospital||No data submitted for December.|
|Mercy Hospital Buffalo||Low discharges for January.|
|Mount Vernon Hospital||No data submitted for December.|
|Myers Community Hospital||No data submitted for July thru September.|
Low discharges for October thru December.
|Nassau University Medical Center||Low discharges for October and November.|
|Newark-Wayne Community Hospital||Low discharges for April, June thru August.|
|NY United Hospital Med Ctr||Low discharges for February and March.|
|Peninsula Hospital||Low discharges for January thru March.|
|Rochester General Hospital||Low discharges for November.|
|Southampton Hospital||No data submitted for September thru November.|
Low discharges for December.
|St. Francis Hospital Beacon||No data submitted for June thru December.|
|The Hospital||No data submitted for September.|
|Yonkers General Hospital||No data submitted for May and June.|
Variations in hospital discharges between years can sometimes be attributed to reporting pattern changes. The following identifies those facilities who have submitted complete data for 1999 with fewer discharges than they reported in 1998 along with the reason for those differences:
|Park Ridge Hospital - Genesee Street Campus||Merged with St. Mary's Hospital Rochester.|
|St. John's Episcopal Hospital||Closed Effective 09/30/1999.|