Data Quality Documentation for Users: Canadian Joint … · Data Quality Documentation for Users:...

Standards and Data Submission

Data Quality Documentation for Users: Canadian Joint Replacement Registry, 2010–2011 Data

Our VisionBetter data. Better decisions. Healthier Canadians.

Our MandateTo lead the development and maintenance of comprehensive and integrated health information that enables sound policy and effective health system management that improve health and health care.

Our ValuesRespect, Integrity, Collaboration, Excellence, Innovation

Table of Contents Introduction ................................................................................................................................... 1

Purpose of the Report ............................................................................................................. 1 Canadian Joint Replacement Registry .................................................................................... 1 Dissemination .......................................................................................................................... 2 Users ....................................................................................................................................... 2

1 Concepts and Definitions ........................................................................................................ 3 1.1 Mandate/Purpose ........................................................................................................... 3 1.2 Core Data Elements and Concepts ................................................................................ 3 1.3 CJRR Populations and Frame........................................................................................ 4

2 Data Limitations ...................................................................................................................... 4 2.1 Data Coverage ............................................................................................................... 4

2.1.1 Over-Coverage ................................................................................................... 5 2.1.2 Under-Coverage ................................................................................................. 5

3 Collection and Non-Response ................................................................................................ 7 3.1 Data Collection .............................................................................................................. 7 3.2 Data Quality Control ...................................................................................................... 9 3.3 Non-Response ............................................................................................................... 9

3.3.1 Unit Non-Response ............................................................................................ 9 3.3.2 Item (Partial) Non-Response ........................................................................... 10

3.4 Measurement Error ..................................................................................................... 13 3.5 Issues of Bias and Reliability ....................................................................................... 13

4 Major Methodological Changes from Previous Years ........................................................... 14 4.1 Major Changes Compared With 2009–2010 ............................................................... 14 4.2 Changes Affecting Longitudinal Comparisons ............................................................ 14

5 Revision History .................................................................................................................... 14

6 Contact Information ............................................................................................................... 14

1


Introduction Maintaining and enhancing the quality of incoming data is essential to CIHI’s mandate to produce high-quality health information. CIHI’s Data and Information Quality Program ensures the continued regular improvement of the quality of CIHI’s databases and registries to meet changing and expanding user requirements and expectations.

CIHI’s Data Quality Framework provides a common way to assess data quality across CIHI’s databases and registries; within the framework, the data quality assessment tool helps users identify priority issues for quality by assessing databases and registries along the following five broad dimensions: accuracy, comparability, timeliness, usability and relevance. The framework’s implementation is part of the larger quality cycle in which problems are identified, addressed, documented and reviewed on a regular basis. Using the framework also standardizes information on data quality and helps to identify priority issues, which in turn leads to continuous improvements. The assessment tool highlights strengths (processes that work well) and identifies areas where existing practices can be improved.

The purpose of this document is to provide information from the user’s perspective to assess the fitness for use of the data.

Purpose of the Report The primary intent of this document is to provide users with sufficient information to assess whether the quality of the information presented by the Canadian Joint Replacement Registry (CJRR) fits their intended use. This document contains information on coverage, data limitations, comparability, major changes and revisions and their impact. It is restricted to the known limitations of CJRR for 2010–2011 (that is, surgery dates from April 1, 2010, to March 31, 2011).

Canadian Joint Replacement Registry CJRR is a Canada-wide registry that collects clinical, demographic, administrative and prostheses information on hip and knee replacement procedures performed in Canada.

The registry was formed as a collaborative effort between CIHI and orthopedic surgeons of Canada. The goals of the registry are to collect, process and analyze data on hip and knee replacements performed in Canada; to support evidence-based decision-making to improve the quality of care for joint replacement recipients; and to conduct analyses pertaining to orthopedic devices and surgical techniques. The hip and knee data collection forms and a list of the registry’s main data elements can be downloaded from CIHI’s website at www.cihi.ca/cjrr.

2


For the 2010–2011 data year, CJRR data was submitted primarily by orthopedic surgeons on a voluntary basis after obtaining consent from patients prior to or at the time of surgery. As of April 1, 2012, all Ontario hospitals have been mandated by the Ontario Ministry of Health and Long-Term Care to submit CJRR data. As of July 1, all health authorities in British Columbia have been mandated by the Ministry of Health to submit data for all hip and knee procedures.

Currently, CJRR supports three different modes of data submission:

Electronic submission from surgeons’ coordinators or vendors;

Web submission through the CJRR Web-Based Data Submission and Reports Tool (as of 2009–2010); and

Paper data collection forms.

Paper forms are entered directly into the database via an internal web-based interface by CJRR data entry staff at the CIHI office in Toronto, Ontario. Electronic data is submitted directly by the physicians’ coordinators, hospitals or third-party vendors based on pre-defined CJRR specifications. The CJRR Web-Based Data Submission and Reports Tool is a data submission application that enables surgeons or their designated staff to submit data electronically to CJRR through a secure internet connection. The tool, available in both English and French, also allows users to run and view surgeon-specific summary reports.

For the 2010–2011 reporting period, a total of 39,486 records were received, of which 10,627 (26.9%) were submitted electronically, 1,019 (2.6%) were submitted through the web-based application and 27,840 (70.5%) were submitted via paper forms.

Dissemination Annual statistics are produced by the registry to provide epidemiological, clinical and surgical analyses on hip and knee replacement procedures across Canada.

The CJRR Comparative Reports are individual reports that are made available to only the surgeon who performed the joint replacement procedure and submitted data to CJRR. Their objective is to provide surgeons with comparable indicators to support performance measurement of and quality improvement in joint replacement procedures.

The registry also responds to data requests from participating surgeons, governments, researchers and other stakeholders. These data requests require conducting new analyses on surgical or clinical data, in the form of aggregate-level data and/or graphical summaries, or by record-level data extractions. All data requests are responded to in adherence with CIHI’s privacy and confidentiality guidelines.

Users Primary users of CJRR data include orthopedic surgeons, health policy-makers and health care administrators. Secondary users include allied health care clinicians, researchers and the general public.

3


1 Concepts and Definitions 1.1 Mandate/Purpose CJRR collects data from across Canada for hip and knee replacements performed in the country. Through CJRR, CIHI provides information so orthopedic surgeons and health care managers can evaluate clinical procedures and monitor the progress of joint replacement recipients over time. Figure 1 shows the flow of data into CJRR for the 2010–2011 data year.

Figure 1: Canadian Joint Replacement Registry Data Flow Diagram, 2010–2011

1.2 Core Data Elements and Concepts

Currently, CJRR collects information on hip and knee replacements only and includes information on patient demographics, type of replacement (primary or revision procedure), surgical approach, fixation mode and implant type. Primary (initial) and revision procedures—mostly elective—are included for the purposes of analysis and reporting.

Note that in CJRR, bilateral procedures are recorded as two separate procedures, one for each side (left and right), which may differ from other CIHI data holdings.

In terms of CJRR data elements, CJRR implemented a new minimum data set (MDS) as of April 1, 2012, aligned with standards established by the International Society of Arthroplasty Registries (ISAR). In 2010–2011, the longer CJRR form was applied. This list of data elements can be found on CJRR’s website (www.cihi.ca/cjrr).

1

2

Surgeon/OR Staff

Surgeon/OR Staff

PaperForms

E-Files Joint ReplacementSurgical and Clinical Data

Canadian Joint Replacement Registry

Annual ReportsAnalyses in Brief Ad-Hoc Requests

Additional Joint ReplacementData From CIHI’s Hospital

Morbidity Database (HMDB) and Discharge Abstract

Database (DAD)

Surgeon/OR Staff3 Web Application

4


1.3 CJRR Populations and Frame The population of interest is all hip and knee replacement procedures performed by all surgeons in Canada, in acute or private facilities.

The population of reference is all hip and knee replacement procedures that were performed in acute facilities in Canada between April 1, 2010, and March 31, 2011, by surgeons who are eligible to participate in CJRR. Surgeons who perform hip or knee replacement procedures are considered eligible. CJRR uses the Discharge Abstract Database (DAD) or the Hospital Morbidity Database (HMDB) as references. CJRR conducts regular comparisons with these two administrative databases to ensure that the frame and reference populations are aligned.

For the purpose of this data quality assessment, the frame units are CJRR-registered orthopedic surgeons who perform hip and knee replacement procedures, and the units of analysis are those orthopedic surgeons as well as the joint replacement procedures submitted to CJRR. The reference period is 2010–2011 (April 1, 2010, to March 31, 2011).

2 Data Limitations This section discusses the issues users should note when interpreting and using CJRR data.

2.1 Data Coverage The HMDB is used as the reference against which CJRR is compared. The exact population of reference is difficult to quantify, as surgeon-level information is not captured in the HMDB. Also, since private facilities are not found in the HMDB, there is no frame available for private facilities.

The HMDB captures a finite set of administrative, clinical and demographic information on hospital inpatient events, including primary and revision joint replacement procedures. It and the DAD are recognized as the only databases that are comparable to CJRR for validation purposes. Using 2010–2011 data, the distribution of various elements was compared between CJRR and the HMDB, including patient age, gender, postal code of residence, type of joint replacement (type of intervention) and reason for primary joint replacement. The data distribution across these elements was found to be comparable between the two databases. Other CJRR elements that are comparable to those in the HMDB are health card number, health card issuing province, procedure date and date of birth.

In the following sections, coverage will be measured in two ways:

1. Comparison by procedure: number of procedures submitted to CJRR compared with the number of procedures in the HMDB.

2. Comparison by surgeon participation, based on estimates only.

5


2.1.1 Over-Coverage Procedure Over-Coverage

In terms of joint replacement procedures reported to CJRR, there is a small potential for over-coverage through duplicate records in the CJRR database. Duplicate entries result when the data provider submits multiple records for a single procedure. Edit checks were implemented in the new relational CJRR database (effective April 2005) to prevent duplicate entries. Exact patient or procedure matches are rejected by the database; however, similar records are allowed. These are then flagged as potential duplicates in data quality reports. As well, since private facilities are not in our population of reference but in CJRR, procedures performed in private facilities will be considered over-coverage.

The over-coverage rate is a minor issue for CJRR, with less than 0.5% for potential duplicate records and just over 2% for private facility procedures.

Surgeon Over-Coverage

Surgeon over-coverage is not a major issue in the CJRR database. Prior to submitting data, all surgeons must first register with CJRR. Only practising orthopedic surgeons are included in the population of reference. Further, information on surgeons who move, retire or cease to perform hip or knee replacements is tracked, once reported to CJRR.

2.1.2 Under-Coverage Procedure Under-Coverage

This is a significant issue for CJRR. There are a few sources of procedure under-coverage in CJRR. First, not all orthopedic surgeons across Canada participate in CJRR (a major source of under-coverage) and the surgeons who are participating may not necessarily submit data to CJRR for all of the procedures they have performed. Further, if patient consent is not provided, their procedures are not captured in the database, resulting in further under-coverage. Therefore, only a portion of all joint replacement procedures captured by the HMDB (which captures all acute care records) are represented in CJRR. Note that as of April 1, 2012, with the implementation of the CJRR MDS, patient consent is no longer being tracked by CJRR.

It should also be noted that there is also some under-coverage within the HMDB, the reference source. Private facilities can report to CJRR but do not report to the HMDB, resulting in under-coverage of HMDB.

6


Tables 1 and 2 present hip and knee replacement coverage by jurisdiction for 2010–2011, comparing CJRR to HMDB. Overall, for procedures related to both joints, the under-coverage rate was approximately 59%.

Table 1: Hip Replacements in CJRR as a Percentage of the HMDB, by Jurisdiction, 2010–2011

Jurisdiction Hip Replacements

in CJRR

Hip Replacements in CJRR

(Excluding Private Facilities)

Hip Replacements in HMDB

Percentage of Hip Replacements in CJRR Found in

HMDB British Columbia 2,739 2,738 6,397 43%

Alberta 2,303 2,086 4,467 47%

Saskatchewan 905 305 1,615 19%

Manitoba 1,557 1,557 1,893 82%

Ontario 3,643 3,643 17,032 21%

Quebec 3,645 3,578 7,968 45%

New Brunswick 714 714 1,127 63%

Nova Scotia 1,196 1,196 1,429 84%

Prince Edward Island 0 0 178 0%

Newfoundland and Labrador 114 114 582 20%

Territories* 8 8 25 32%

Total 16,824 15,939 42,713 37%

Note * Territories include Yukon, the Northwest Territories and Nunavut.

Table 2: Knee Replacements in CJRR as a Percentage of the HMDB, by Jurisdiction, 2010–2011

Jurisdiction

Knee Replacements in

CJRR

Knee Replacements in CJRR (Excluding Private Facilities)

Knee Replacements in

HMDB

Percentage of Knee

Replacements in CJRR Found in

HMDB British Columbia 3,522 3,521 6,886 51%

Alberta 3,174 2,915 5,053 58%

Saskatchewan 1,573 1,573 1,889 83%

Manitoba 1,940 1,940 2,058 94%

Ontario 4,968 4,971 21,832 23%

Quebec 4,546 4,457 9,049 49%

New Brunswick 1,092 1,092 1,223 89%

Nova Scotia 1,562 1,562 1,751 89%

Prince Edward Island 0 0 202 0%

Newfoundland and Labrador 253 253 727 35%

Territories* 32 32 63 51%

Total 22,662 22,316 50,733 44%

Note * Territories include Yukon, the Northwest Territories and Nunavut.

7


Surgeon Under-Coverage

The surgeon participation rate is difficult to ascertain, as there is no definitive source that lists the number of orthopedic surgeons in Canada who currently perform hip and knee replacement procedures. However, with the assistance of the CJRR Advisory Committee and the Canadian Orthopaedic Association, CJRR estimates that the surgeon under-coverage rate is approximately 45%.

3 Collection and Non-Response This section provides a summary of the data collection, processing and quality control activities undertaken by CJRR and how they may affect an external user’s analysis.

3.1 Data Collection The following procedures and practices were in place.

Paper Forms

CJRR staff provided clients with regular support and advice regarding paper submissions.

Prepaid, self-addressed courier envelopes were supplied to CJRR’s participating surgeons to facilitate the return of completed data collection forms in a secure manner.

A comprehensive micro-editing process took place prior to data entry to improve the ease of data entry for internal CJRR staff.

CJRR staff conducted follow-up via phone for paper forms with incomplete or inconsistent information; however, not all responses were available from the surgeons’ offices.

CJRR newsletters or communications were released to clients to communicate new processes or suggestions to improve data quality.

Electronic Submissions

CJRR released an updated e-submission specification document (Canadian Joint Replacement Registry Electronic Data Submission Requirements), which outlined changes to the database structure and specifications. This document was made available to all vendors who have signed the vendor licensing agreement.

The aforementioned specification document was circulated to all submitting vendor groups per CIHI’s standards, ensuring that lead time for clients was accommodated.

CJRR provided a Data Submission Error Report after each data submission was completed, which outlined the result of the data submission and any errors.

Teleconferences with electronic data vendors took place to improve timeliness of data submission and to answer any questions or concerns.

8


Web Submissions

CJRR staff provided regular support to web submitters.

The web tool applies validations at point of entry. Therefore, only data that passes the checks can be saved in the database.

With the implementation of the MDS in 2012–2013, CJRR aims to further reduce the burden of record submission by significantly paring down the number of data elements and thus the time required to prepare data for submission.

The following CJRR practices were in place to encourage new and ongoing participation and to enhance awareness of the registry:

Professional development credits were offered as an incentive to CJRR participating surgeons, in collaboration with the Royal College of Physicians and Surgeons of Canada. One credit was issued for every six submitted records, for a maximum of 15 credits per calendar year.

CJRR released analytical products, at no cost to clients, to encourage their continued contribution to CJRR.

Presentations by CJRR at the Canadian Orthopaedic Association and at other provincial-level orthopedic conferences encouraged CJRR-eligible, non-participating surgeons to sign up with the registry.

CJRR worked collaboratively with its Advisory Committee to foster consensus and build on strategies and new business.

CJRR staff responded to data requests from participating surgeons or data suppliers via centralized email. Ad-hoc requests were acknowledged within 48 hours in accordance with corporate service time frames.

Within the constraints of privacy and confidentiality guidelines, surgeons were able to request an electronic copy of their own data to assist with reviews of their surgical practices.

CJRR produced specialized studies for the purpose of research and presentation at professional conferences and symposia.

CJRR produced and distributed Surgeon Comparative Reports, a tool that allowed surgeons to conduct peer-group comparisons.

9


3.2 Data Quality Control The following quality control measures were applied to CJRR data:

Verification of samples of paper forms was completed for the 2010–2011 data. As a measure of internal consistency, this task was performed by a staff member who did not enter the original paper form data. Data entered was matched with verification guidelines to ensure consistent data entry.

For paper forms, the raw data was entered into the CJRR database according to pre-established data entry rules.

Data entered into all fields in the data entry program was subjected to automated logic edits to ensure completion of mandatory fields and logic conditions.

Two types of errors were generated if a record failed an edit check:

– Warnings occurred when optional data elements were missing from a record. An error log was produced at the end of the record displaying each field that had the error and the description of the error.

– Severe errors occurred when mandatory data elements were left blank or the format of the data was invalid. Records with severe errors could not be saved in the database.

Each surgeon was assigned a unique identification number to prevent creating duplicate surgeons in the database.

Data quality reports were produced to identify potential duplicate patient and/or procedure records, as discussed in Section 2.1.1. True duplicates were then deleted.

3.3 Non-Response 3.3.1 Unit Non-Response 3.3.1.1 Procedure Unit Non-Response

It is difficult to ascertain the procedure unit non-response rate, since our population of reference (HMDB) does not provide information at the surgeon level. As a result, we cannot determine the number (or percentage) of procedures performed by each CJRR-participating surgeon that were not submitted to CJRR.

10


3.3.1.2 Surgeon Unit Non-Response

In CJRR, the surgeon unit non-response rate can be defined as the number of CJRR-participating surgeons who did not submit data in 2010–2011 (the numerator) divided by the total number of registered and active surgeons in CJRR (the denominator). Table 3 shows the number of active surgeons who submitted data in 2010–2011 as a percentage of those who are currently registered and active in CJRR. The overall surgeon unit response rate was 82%.

Table 3: Participating Surgeon Response Rate, by Jurisdiction, CJRR, 2010–2011

Jurisdiction of Participating Surgeon

Number of Surgeons Who

Submitted to CJRR During the

Reporting Period

Number of Registered and

Active Surgeons in CJRR*

Unit Response Rate

British Columbia 59 84 70%

Alberta 55 58 95%

Saskatchewan 27 28 96%

Manitoba 35 35 100%

Ontario† 51 85 60%

Quebec 110 135 81%

New Brunswick 26 27 96%

Nova Scotia‡ 30 29 100%

Prince Edward Island 0 0 N/A

Newfoundland and Labrador 8 11 73%

Territories§ 2 2 100%

Total 403 494 82%

Note * Surgeon’s status as of the end of the reporting period. † One facility in Ontario submits data without surgeon information; thus, surgeons practising in that facility

have not been included in the totals. ‡ Two surgeons from Nova Scotia who submitted data in 2010–2011 had a change in status (to inactive) as

of the end of the reporting period. § Territories include Yukon, the Northwest Territories and Nunavut.

3.3.2 Item (Partial) Non-Response With the introduction of mandatory fields in the relational CJRR database at the beginning of April 2005, the completeness of the database improved significantly. And with the implementation of the MDS in April 2012, for which all core data elements are mandatory, the item (partial) non-response rate will be further reduced.

The current CJRR database was designed for 100% completion of all mandatory data elements. Electronic and web submissions cannot be saved in the CJRR database if mandatory data elements are missing. Follow-up is done on mandatory data elements that have been left blank on the paper forms submitted; the information is subsequently verified by the corresponding surgeon. There were 1,801 forms (6.5%) with missing mandatory data elements in 2010–2011.

11


The optional data elements in the CJRR relational database can be grouped into three categories; the item non-response rate for optional data elements in each category is shown in tables 4 to 7. Note that the accuracy of the response was not assessed. Data from two facilities in Ontario that submit separate data files to CJRR that do not conform to CJRR specifications was not included.

a) Demographics and Administrative Information

Table 4: Item Non-Response Rate for Optional Data Elements (Demographics and Administrative Information), 2010–2011

Optional Data Element

Hip Procedures Knee Procedures Number of

Records With Missing Data

Element

Total Number of Records

Item Non-Response

Rate

Number of Records With Missing Data

Element


Item Non-Response

Rate

Referral Date 12,803 15,543 82% 16,242 20,703 78%

Date of First Consult 11,503 15,543 74% 14,946 20,703 72%

Date of Decision for Surgery

7,321 15,543 47% 8,338 20,703 40%

Overall 68% 64%

b) Surgical and Clinical Information

Table 5: Item Non-Response Rate for Optional Data Elements (Surgical and Clinical Information), 2010–2011


Hip Procedures Knee Procedures Number of

Records With Missing Data

Element


Item Non-Response

Rate

Number of Records With Missing Data

Element


Item Non-Response

Rate

Patient Height 5,960 15,543 38% 6,523 20,703 32%

Patient Weight 5,802 15,543 37% 6,325 20,703 31%

Surgical Approach 69 15,543 0% 82 20,703 0%

Antibiotic Use (Flag) 571 15,543 4% 191 20,703 1%

Antibiotic Duration 49 14,940 0% 544 20,467 3%

Minimally Invasive Surgery (Flag)

377 15,543 2% 617 20,703 3%

Deep Vein Thrombosis Prevention (Flag)

9 15,543 0% 4 20,703 0%

OR Environment 131 15,543 1% 174 20,703 1%

Overall 9% 8%

12


c) Parts and Prostheses Information

Collecting parts and prostheses information is an important function of CJRR. However, in the 2010–2011 data year, all relevant fields were optional given the variety of prosthesis combinations possible. Therefore, it was difficult to ascertain whether missing information was truly missing or just not applicable. Tables 6 (for hip procedures) and 7 (for knee procedures) show the item non-response rate for parts and prostheses information. The denominator for each item non-response rate was the total number of records related to the replacement of the specified part (Optional Data Elements). The numerator was the number of records with at least one element missing (manufacturer, catalogue number or lot number).

Table 6: Item Non-Response Rate for Optional Hip Procedure Data Elements (Parts and Prostheses Information), 2010–2011


Number of Records With at Least One Element

Missing (Manufacturer, Catalogue or Lot Number)


Item Non-Response Rate

Femoral Component 1,248 14,499 9% Femoral Head 212 15,429 1% Acetabular Component 800 14,088 6% Acetabular Insert/Liner 1,032 13,643 8% Ring/Cage 4 47 9% Cement 265 2,772 10%

Overall 6%

Table 7: Item Non-Response Rate for Optional Knee Procedure Data Elements

(Parts and Prostheses Information), 2010–2011


Number of Records With at Least One Element

Missing (Manufacturer, Catalogue or Lot Number)


Item Non-Response Rate

Femoral Component 266 20,286 1% Tibial Component 1,395 20,485 7% Patellar Component 342 13,627 3% Modular Stem* 92 876 11%

Overall 6%

Note * Includes both femoral stem and tibial stem. A record is deemed to have missing information if a modular

stem was used (either femoral or tibial), and neither femoral nor tibial stem information (manufacturer, catalogue number and lot number) was complete.

Paper forms with missing mandatory information were followed up by telephone, and corrections were made based on feedback received from surgeons and/or their offices. However, not all information was obtained this way.

13


3.4 Measurement Error The complex design of the CJRR database prevents the calculation of an overall error rate. However, more than 50% of the data elements are mandatory, and strong edit checks are incorporated into the application to hedge the risk of error. These edit checks include logic checks to ensure that related variables are answered appropriately, resulting in an increase in CJRR data quality.

3.5 Issues of Bias and Reliability The extent of bias in CJRR is unknown. In particular, procedure under-coverage and surgeon under-coverage are major sources of bias. However, these biases are difficult to evaluate. With mandated participation across Ontario and British Columbia effective 2012–2013, selection bias in these two provinces will be essentially eliminated.

There may be some degree of inconsistency due to differing clinical definitions. For example, what is considered a revision procedure by one surgeon may not be considered as such by another. Inconsistencies may also occur due to differences in defining techniques or procedures (for example, defining minimally invasive surgery in relation to body size or defining a procedure as bilateral).

All variables in the 2010–2011 data set were subjected to CJRR database validity, logic, range and consistency checks. Validity rules pertaining to electronic submission specifications were documented in the CJRR electronic submission specifications for 2010–2011. Data is checked against an accepted set of responses as well as proper formatting. Electronic submission data that does not pass these checks is flagged as erroneous, not saved in the database and sent back to the data supplier for correction. Paper forms that fail validity checks are set aside for follow-up with the data provider.

Further, all core (mandatory) variables in the CJRR data set were checked for validity using SAS programs before these core variables were used for any analyses.

Records containing patients outside of pre-defined age and weight ranges were flagged for follow-up. Records containing admission and surgery dates that were out of chronological sequence were manually verified, and the appropriate corrections were applied once clarification was obtained from the surgeon’s office.

14


4 Major Methodological Changes from Previous Years

4.1 Major Changes Compared With 2009–2010 No major changes were made between 2009–2010 (the data assessed in the previous Data Quality Documentation for Users report) and 2010–2011.

4.2 Changes Affecting Longitudinal Comparisons No changes have been made to the CJRR data collection forms since 2009–2010. However, significant changes were made as of April 1, 2012, for 2012–2013. Based on consultation with and input from the CJRR Advisory Committee and members of the Canadian Orthopaedic Association, CJRR implemented the MDS for hip and knee replacements. Work has been completed to align the data currently collected by CJRR with the MDS recommended by the ISAR. The MDS will significantly reduce the burden on data providers while retaining key data elements. The usefulness of the data will be greatly improved once CJRR identifies a sustainable way of accessing the comprehensive product information that can be leveraged in the implant product barcodes (that is, manufacturer, catalogue number and lot number), a project of interest to CJRR.

5 Revision History CJRR data is subject to revision if any corrections are received from the data providers. CJRR continues to accept data beyond the reporting period deadline; thus there may be slight variations in reported data over time.

Since CJRR uses surgery date for its definition of fiscal year, data from the current fiscal year, as well as previous fiscal years, is received throughout the year and entered into the CJRR database. As a result, data reported in earlier analytical reports may be changed by late data submission. For example, if the patient’s surgery was performed in 2009–2010 but the record was not submitted to CJRR until 2010–2011, the record would not have been included in the 2009–2010 data cut, as data is grouped by fiscal year (that is, surgery date).

6 Contact Information For more information, please contact the CJRR team at [email protected].

Production of this report is made possible by financial contributions from Health Canada and provincial and territorial governments. The views expressed herein do not necessarily represent the views of Health Canada or any provincial or territorial government.

All rights reserved.

The contents of this publication may be reproduced unaltered, in whole or in part and by any means, solely for non-commercial purposes, provided that the Canadian Institute for Health Information is properly and fully acknowledged as the copyright owner. Any reproduction or use of this publication or its contents for any commercial purpose requires the prior written authorization of the Canadian Institute for Health Information. Reproduction or use that suggests endorsement by, or affiliation with, the Canadian Institute for Health Information is prohibited.

For permission or information, please contact CIHI:

Canadian Institute for Health Information495 Richmond Road, Suite 600Ottawa, Ontario K2A 4H6

Phone: 613-241-7860Fax: [email protected]

© 2012 Canadian Institute for Health Information

Cette publication est aussi disponible en français sous le titre Documentation sur les données à l’intention des utilisateurs : Registre canadien des remplacements articulaires, données de 2010-2011.

www.cihi.caAt the heart of data

Talk to UsCIHI Ottawa495 Richmond Road, Suite 600Ottawa, Ontario K2A 4H6Phone: 613-241-7860

CIHI Toronto4110 Yonge Street, Suite 300Toronto, Ontario M2P 2B7Phone: 416-481-2002

CIHI Victoria 880 Douglas Street, Suite 600 Victoria, British Columbia V8W 2B7 Phone: 250-220-4100

CIHI Montréal1010 Sherbrooke Street West, Suite 300Montréal, Quebec H3A 2R7Phone: 514-842-2226

CIHI St. John’s140 Water Street, Suite 701St. John’s, Newfoundland and Labrador A1C 6H6Phone: 709-576-7006

Data Quality Documentation for Users: Canadian Joint … · Data Quality Documentation for Users:...

Documents

Transcript of Data Quality Documentation for Users: Canadian Joint … · Data Quality Documentation for Users:...