Skip to main content

Table 1 Data Sources and Methods Used to Create Study Measures

From: How well does the minimum data set measure healthcare use? a validation study

Study Measures

Administrative Health Care Use Records

InterRAI (MDS) Assessments

NURSING HOME (NH) USER STATUS

Source

Method

Source

Method

i) Residents who were Newly Admitted During the Study Period (versus not)

NH Utilization File

This file was reviewed retrospectively until 1984, to identify residents admitted during the study period who had not been previously admitted to a NH.

Admission Record

The first admission record was selected for each person. People were identified as a newly admitted NH resident if item: i) AB2a (denotes where the person was admitted from) ≠ 4 (denotes a 24-h nursing care facility); or ii) AB2A = 4 & AB2b (the facility # the person was admitted from) ≠ a Manitoba-based NH ID.

ii) Residents Newly Admitted into a NH Directly from Hospital vs Another Location (calculated on people defined as newly admitted in both data systems)

Hospital Discharge Abstract Database (DAD) linked to the NH Utilization File

DAD hospital discharge dates were aligned with NH admission dates. People transferring into NH directly from hospital had ≤1 day between these dates.

Admission Record

People were defined as transferring directly from hospital if item AB2a on this record was coded as any of 1 (inpatient acute care), 2 (inpatient rehab), 5 (inpatient psychiatry) or 7 (inpatient specialized rehab).

iii) Residents with 1+ NH Transfer During the Study Period (versus not)

NH Utilization File

This file contains a NH ID for each resident. Residents who transferred NHs had multiple NH IDs in this file, and the date of the ID change was also captured.

Admission Record

Residents were defined as having transferred NHs if: i) item AB2a = 4 (resident arrived from another NH); or ii) AB2b (facility # admitted from) contained a Manitoba NH ID which did ≠ the present facility ID.

iv) Residents who Died (versus who did not) During the Study Period

Registry File

This file contains a unique identifier for every Manitoban. The Insurance Cancellation Code in this file was used to define the date of each person’s death.

Discharge Record

The last discharge record was selected for each person. People were identified as dying if item i) R3a = 11 (denotes people who were deceased); or ii) AA8 = 6 (discharged and not likely to return) and R3a (discharge locations) was 1 (inpatient acute care), 2 (inpatient rehab), 5 (inpatient psychiatry) or 7 (inpatient specialized rehab).

HEALTH CARE USE

    

i) In-patient Hospitalization (yes vs no, and frequency of visits for people with 1+ hospitalization in each data system).

Hospital Discharge Abstract Database (DAD)

DAD was used to count the number of times people were hospitalized with a length of stay > 1 day, overall and within each MDS episode.

MDS full assessments

Item P5 records the number of times each person was admitted to a hospital in the previous 90 days. This 90-day period was defined as the MDS episode.

ii) Emergency Department Visits not ending in Hospitalization (yes vs no, and frequency of visits for people with 1+ ED visit in each data system).

The Emergency Department Information System (EDIS)

This file was used to count the number of times each person visited an ED, overall and within each MDS episode. Visits ending in hospitalization were excluded using the EDIS disposition code.

MDS full assessments

Item P6 records the number of times each person visited the ED without being hospitalization in the previous 90 days. This 90-day period was defined as the MDS episode.

iii) Number of Days the Resident was Examined by a Physician (0 versus 1+ days; and frequency of days for people with 1+ examination in each data system).

The Medical Claims File*

This file was used to count the number of days people who had an ambulatory care physician visit (e.g., where the patient was cared for in the NH), overall and within each MDS episode. Ambulatory care visits are defined in a table footnote.

MDS full and quarterly assessments

Item P7 records the number of days each person was examined by a physician (or authorized professional such as a nurse practitioner) in the previous 14 days. This 14-day period was defined as the MDS episode.

  1. *Prefix ‘7’ tariff codes denote ambulatory care physician visits in medical claims. Tariff code 8511 (‘general scheduled visit for chronic care’) accounts for 77.7% of all ambulatory care physician visits measured in the study period, code 8513 (‘regular visit for patients aged 70 years and older’) accounts for 9.1% of all such visits, and tariff code 8500 (‘complete physical exam for patients aged 70 years and older’) accounts for 3.4% of all ambulatory care physician visits measured in the study period