Verification of Laboratory Information System - An Observational Study

Introduction

The laboratory information system (LIS) in a clinical laboratory is used for the collection, processing, storage, and retrieval of the data that has been produced in the clinical laboratory.¹ To ensure the required quality in healthcare laboratory reporting, it is essential for the laboratory to have a proper LIS in place. The LIS integrated with the analyzers should be verified in a proper manner to run the laboratory at its optimal level.² A hospital information system (HIS) is a broad, integrated system designed to manage the administrative, clinical, and financial aspects of the hospital.³ Some of the secondary and tertiary care hospitals will preferably have the same HIS and LIS providers, so they can perform to their optimal level and reduce any kind of errors occurring.

A properly verified LIS will always be effective in obtaining patient results on time and reduces unnecessary use of any other paper works affecting the turnaround time of the patient results.⁴ LIS offers to reduce the human errors involved in the reporting process and also increases the accuracy of the data and storage. The LIS also offers longer and safer data storage when compared to traditional document keeping. So, it is very important to verify the LIS, which is used in every clinical laboratory as per the accreditation requirements.

The three phases of laboratory processes are pre-analytical, analytical and post-analytical phases and the LIS plays a vital role in all the three processes, releasing timely quality report. The pre-analytical phase involves collection and storage of data related to the patient, such as patient’s name, age, and gender, assigning a patient with a unique identification number and also giving each sample the time of collection, and the time of sample reporting. In the analytical phase, the interfaced LIS will allow the analyzer to identify the sample based on its sample identification number and the requested tests for that particular sample. In the post-analytical phase, LIS helps in the transfer of sample reports to the end users.

There is a paucity of research on the verification of LIS, especially in the recent times. Our study intended to verify the LIS by assessing the accuracy of data transfer between autoanalyzers and the LIS.

Materials and Methods

The study was conducted for one year after obtaining the institutional ethical clearance (FMMC/FMIEC/255/ 2023, dated 26.04.2023). This study was conducted in the Biochemistry central laboratory of a tertiary care hospital. It was a prospective, observational study of the LIS interfaced with autoanalyzers. According to the laboratory’s policy, for a very large laboratory, minimum of 1% daily samples are considered for checking the accuracy of data transfer. Therefore,

a) Before connecting automated laboratory conveyer system to the autoanalyzers - 32 samples [16 samples each for two autoanalyzers, among which eight- Inpatient (IP) and eight-Outpatient (OP)].

b) After connecting automated laboratory conveyer system to the autoanalyzers - 32 samples (16 samples each for two autoanalyzers, among which eight-IP and eight -OP).

c) After installation of new LIS - 32 samples (16 samples each for two autoanalyzers, among which eight -IP and eight -OP).

d) After three months of new LIS establishment - 32 samples (16 samples each for two autoanalyzers, among which eight -IP and eight -OP).

e) For the follow-up of the corrective action taken - 14 samples, which had flag symbols (< & >) in the autoanalyzers, were selected and verified.

Thus, a total of 142 samples were considered for the study.

Inclusion and exclusion criteria

Samples with a minimum of five parameters in a report, transferred from analyzers having bidirectional interfacing with LIS and HIS were included. Analyzers which were not interfaced were excluded. The parameters not having the bidirectional interface were also excluded for the study.

Data collection

Data were collected as follows,

Data from test request form (TRF) which had the requested tests from the clinicians were noted.

Data from the analyzer - Instrument output of the patient result was collected.

Data from the LIS - To assess the accuracy of data transfer, data were collected from three different places.

a) Result entry and updating end by clerical staff.

b) Authorization end by laboratory consultants.

c) Report dispatch end in outpatient and inpatient departments.

At each step of the data collection, the patient’s demographic data, values of test results, units of measurements, method name, consultant signature, were verified for different tests. For the non-conformance in the data transfer, root cause analysis by “why-why analysis method” was done and corrective and preventive actions were taken.

Phases of data collection

Data collection was done in different phases of labo-ratory upgradation, such as:

a) Before connecting automated laboratory conveyer system to the autoanalyzers.

b) After connecting automated laboratory conveyer system to the autoanalyzers.

c) After installation of new LIS.

d) After three months of the new LIS installation.

e) Follow up of the corrective action taken.

All samples were subjected for surveillance of data integrity and Laboratory information system verification at various stages according to the checklist. Accuracy of data transfer from analyzer to LIS was calculated in terms of percentage.

Results

The data from a total of 128 samples were collected from two autoanalyzers. 127 samples had accurate data transfer, while one sample showed error in data transfer. So, the accuracy of data transfer from the two autoanalyzers to the LIS was 99.2%. The Nonconformance (NC) was 0.8%. Corrective and preventive actions taken for the NC are listed in Table 1.

After taking the preventive action, next 14 samples were verified, and all the samples showed accuracy in the data transfer with no errors observed.

Discussion

LIS plays a major role in day-to-day processes of the laboratory by managing all that data related to patient tests and results. The current study attempted to verify the LIS for accurate data transfer from the autoanalyzer to the LIS through different phases of laboratory upgradation which included, change in the autoanalyzer, change in the LIS, connecting automated sample conveyer system to the autoanalyzer and after three months of new LIS installation. A total of 142 sample data were collected for checking accurate transfer of test results from autoanalyzer to LIS. This study not only focused on the accuracy of results transfer from autoanalyzer, but also verified other demographic details of each patient, such as if the name, age, identification numbers were properly mentioned, the timings of sample collection, transport, reception and reporting, etc. Even the result retrieval after a certain period of time was checked.

Our study used ‘yes/no’ type checklist which was standardized by the quality management system of the laboratory for the verification of LIS. The checklist was included in the quality system procedure of the laboratory. For root cause analysis, “why-why analysis” methodology was used and one non-conformance incident in the accuracy of result transfer was identified.

Results of the study indicated a 99.2% accuracy in data transfer from autoanalyzer to the LIS. Previously, a study conducted by Biljak et al., on the validation of laboratory system reported a 99.5% accuracy in data transfer from the autoanalyzer to the LIS, indicating that the verified LIS and the autoanalyzers performed well in the patient data management.²

A study by Di Jin et al., on “Development and implementation of an LIS-based validation system for auto verification toward zero defects in the automated reporting of laboratory test results” recognized errors such as specific warning target errors, algorithm code error, software compatibility problem. They resolved these problems either by adding supplementary algorithm code or by resetting the algorithm code. In our study, similar error with respect to the coding of flag symbols (< & >) not getting interfaced from the autoanalyzer to the LIS was indentified.⁵

A study on validation of customized laboratory information management system by Bund C et al., stated that a laboratory information management system should always focus on user requirement specification and the data entry fields should be available for all the details related to patients in the LIS which are nothing but the functional specifications of LIS, such as patient’s name, age, unique identification number, sample number, time of collection, time of transportation, time of result generation, time of report authorization and time of dispatch. The LIS should be involved in all these processes from the sample test request till the final reporting through sample identification with unique code, test description, result entry and signature/ authorization of test results. In our study, all the above-mentioned user requirement and functional specifications were assessed and verified with 99.2% accuracy.⁶

Limitations of the study included a small sample size, a short study duration, and the inclusion of only two autoanalyzers. Other analyzers in the laboratory were not considered. Future studies could address these limitations by considering larger sample size, longer study duration and including a broader range of instruments interfaced with the LIS. Also, verification of full scope of parameters can be considered for a more comprehensive analysis.

Conclusion

Verification of LIS is an important process in a very large laboratory with high sample load, and the goal of accuracy should be 100%. The LIS and auto analyzers which were verified during different stages of laboratory upgradation showed good performance in the transfer of patient results between LIS and autoanalyzers. This reduces the need for additional manual work such as entering of each patient results from the autoanalyzer to the LIS. The verification and continuous assessment of the LIS should be conducted periodically, and as well as when there are changes in the autoanalyzer, LIS software change, the introduction of newer tests to the menu or changes in the location of the laboratory. Such verification ensures the accuracy of data transfer, which is crucial for accurate clinical decision-making to initiate timely treatment management.

Conflict of interest

Nil