Data validation Policy

Policy Type  Clinical  Administrative Issue Date: 1 September 2021

Review Date: NA
Policy No.: BRFRH/QPS/P – 002
Revision No.: NA
Next Review
Reference Standard QPS.3. 1 September 2023
Date:

1. PURPOSE
1.1. To establish an internal process for data validation in order to evaluate whether data
have been generated according to specifications, satisfy acceptance criteria, and are
appropriate and consistent with their intended use.

1.2. To ensure that data driven assessments and decisions have been made based on
validated and reliable data.

1.3. To define the frame of reference for interpreting the collected data during the
measurement phase.
2. POLICY STATEMENTS
2.1. BRFRH recognizes the importance of high quality data in ensuring continued delivery
of improved patient care and BRFRH is committed to improving data quality
standards.

2.2. BRFRH gathers and stores large amounts of data both manually and electronically,
with increasing reliance being placed on this data to validate delivery of improved
patient care, performance improvement and management decision making. It is
therefore imperative that we achieve and maintain the highest possible data quality
standards whilst remaining focused on the predetermined need for that data.

2.3. Data is collected and processed by many personnel across BRFRH. Data quality may
therefore be affected by a wide range of activities. The need for good quality data
must be embedded in the culture, values and actions of BRFRH personnel. Validation
of the data collected shall ensure a high level of data accuracy and validity.

2.4. Data validation shall be done whenever collected data meet the criteria mentioned
below:

BRFRH/QPS/P – 002 Page 1 of 6

 Data validation Policy
2.4.1. Clinical measures that are intended to help evaluate and improve an important
clinical process or outcome; first two data sets shall be validated.

2.4.2. A change has been made to an existing measure, such as the data collection
tools have changed or the data abstraction process or collector has changed;
first two data sets shall be validated.

2.4.3. The data resulting from an existing measure have changed in an unexplainable
way (data point outside limit); once and then depending on accuracy level.

2.4.4. The data source has changed, such as when part of the patient record has been
turned into an electronic format and thus data source is now both electronic
and paper; first two data sets shall be validated.

2.4.5. The subject of the data collection has changed, such as changes in average age
of patients, comorbidities, research protocol alterations, new practice
guidelines implemented, or new technologies and treatment methodologies
introduced; first two data sets shall be validated.

2.5. Data validation shall be initiated by the department concerned with data collection
or by the Quality Officer.

2.6. A minimum of 90% accuracy level between the first data set and the second data set
shall be required for data to be considered as valid.

2.7. Data validation shall be done through two approaches:

2.7.1. Inter rater reliability: Data collection by a second abstractor comparable in

level of expertise to the first abstractor but not involved in the original data
collection process; using the same data collection tool/approach on a sample
of the original population (i.e. collection of a key library of measure indicator
by two different abstractors using the same data collection tool).

2.7.2. Criterion-related validity: Data collection by the same or different abstractor

using an alternative data collection method on the same population (i.e.
collection of pressure ulcer incidents through near miss/incident reports then
through medical records audit or ICD codes report).

BRFRH/QPS/P – 002 Page 2 of 6

 Data validation Policy
3. SCOPE AND RESPONSIBILITIES
3.1. This policy applies to the validation of administrative and clinical patient related data
collected from BRFRH’s paper and electronic based systems.

3.2. This policy applies to all data collections, including those provided for by regulatory
authorities and statute, held by or within BRFRH. It includes collections of patient
information, corporate, financial and personnel information where one or more of
the following conditions are met:

3.2.1. The data collection is used to meet business, operational or legislative

requirements.

3.2.2. The data collection contains personal information.

3.2.3. The data collection is used for reporting at a state level, national level or
external to the health service where the data collection resides.

3.2.4. The data collection is used across multiple health services.

3.3. The scope of this policy includes both paper based and electronic data.

3.4. The policy applies to all employees of BRFRH, including temporary staff and
contractors.
4. DEFINITIONS/ABBREVIATIONS

4.1. Data Quality: the totality of features and characteristics of data that bears on its
ability to satisfy a given purpose. Data quality can also be defined as the degree to
which a set of characteristics of data fulfills requirements

4.2. Data Collection: is a systematic gathering of data for a particular purpose from various
sources.

4.3. DataValidation: is a systematic process of ensuring that a program operates on clean,

correct and useful data. It uses routines that check for correctness, meaningfulness,
and security of data that are put into the system.

BRFRH/QPS/P – 002 Page 3 of 6

 Data validation Policy
4.4. Data Verification: includes assessing data accuracy, completeness, consistency,
availability, and internal control practices that serve to determine the overall
reliability of the data collected.

4.5. Data Element Agreement Rate (DEAR): is a one-to- one comparison of consensus
between the original abstractor and the re-abstractor's findings at the data element
level, including all clinical and demographic elements.

4.6. Accuracy: is defined as how close to the real value a measurement is.

4.7. Abstractor: refers to the person doing the data collection.

5. PROCEDURES
5.1. The Quality Officer initiates data validation through direct request from the
department head collecting the data or if a need arises for data validation.

5.2. Validation using the inter rater reliability approach:

5.2.1. The concerned department in collaboration with the quality officer decides on
who will be responsible for data re-collection.

5.2.2. When sampling is needed, the following guidelines will be used for selecting a
representative sample size for the second data set out of the original sample
used for the first data set as detailed in the example below:

Quarterly Number of Medical Records Validation Sample size

Originally Abstracted
180 records or greater. At least 5% or a maximum of 50
sampled records.
<180 records. At least 9 sampled records or if <9
records, 100%.

5.2.3. The data will be re-collected by the second abstractor agreed upon.

5.2.4. The Quality Officer will analyse the second data set and compare the results of
the first data set to those of the second data set.

5.2.5. The accuracy level, between the data collected in the first round and the data
collected in the second round, will be calculated by dividing the number of data

BRFRH/QPS/P – 002 Page 4 of 6

 Data validation Policy
elements found to be the same by the total number of data elements and
multiplying that total by 100. The rate is called Data Element Agreement Rate
(DEAR).

5.3. Validation using the criterion based approach:

5.3.1. The concerned department in collaboration with the Quality Officer decides
on the alternative data collection method.

5.3.2. The data will be re-collected by the same or different abstractor as agreed
upon.

5.3.3. The Quality Officer will analyse the second data set and compare the results of
the first data set to those of the second data set.

5.3.4. Accuracy rate will be calculated by dividing the number of matches by the
higher number of cases identified multiplied by 100.

5.4. If the accuracy level is ≥ 90%, the data shall be considered valid and will be approved
and used for decision-making.

5.5. If the accuracy level is found to be below 90%, the reasons for the discrepancy
between the first and the second data sets should be noted and corrective actions
should be taken.

5.6. The corrective actions taken will depend on the identified causes of the discrepancy
between the data sets.

5.7. If the identified causes are related to unclear data definitions or improper data
collection methodologies, then training for the abstractor that collected the first
data set, will be initiated by the Quality Officer.

5.8. If the identified causes are related to fake data, this will be handled through
disciplinary actions, as per BRFRH policies.

5.9. If a new set of data needs to be collected, the Quality Officer along with the
concerned departmental head, will decide on the new abstractor responsible for
collecting the new set.

BRFRH/QPS/P – 002 Page 5 of 6

 Data validation Policy
5.10. A new sample will be re-collected by the identified and trained abstractor and re-
analysed by the Quality Officer.

5.11. The Quality Officer shall continuously ensure that data collected in BRFRH meet
validation requirements, and will remind concerned departments to conduct data
validation.
6. RELATED FORMS AND DOCUMENTS
Nil

7. APPENDICES:
7.1. Document Control

NAME DESIGNATION SIGNATURE DATE

PREPARED BY: 20/08/2021

Ms.Karunamrutha Senior Manager Quality
REVIEWED BY: 30/08/2021
Mr Jayadeep Reddy Unit Head

APPROVED BY: Dr.Prashanth K Group Medical Director 30/08/2021

7.2. Revision History

# Revision Date Change made by Reason for change Clause added/

deleted

7.3. Review History

Review Revision Reviewed by Signature Approved by Signature Next

Date Review

BRFRH/QPS/P – 002 Page 6 of 6