Select Page

Feasibility of Analysis of Lab Result Patterns for Patient Results

Vensa has embarked on this project to enable the analysis of laboratory result patterns that aims for faster and safer dissemination of results to patients.

cute girl sitting in between her grandparents

Key members of the research team and governance are:

  • Samuel Wong – Principal Investigator
  • Dr Tane Taylor – Maori Research and Clinical Lead
  • Dr Jason Hwang – Vensa Global Clinical Advisor
  • Dr Jim Vause – Chair of the Independent Advisory Panel


Project began mid-2019
In Progress

Vensa has embarked on this project to enable the analysis of laboratory result patterns that aims for faster and safer dissemination of results to patients. It addresses the problem of increasing diagnostic laboratory tests being ordered which therefore requires the burden for ongoing clinical interpretation once the results are received to determine appropriate actions. The vast majority of laboratory results ordered are considered normal, but each patient’s “normal” may still suggest underlying needs for treatment or management.  With increasing demand from patients to be empowered with results,  the safe, meaningful and co-ordinated dissemination of these results remains a challenge. 

The ultimate challenge is using data science to determine the clinical reasoning logic in ordering diagnostics tests that is personalised to a patient, so the appropriate pathways are automatically initiated when results have arrived back to the GP, including the filing of these results, and ideally sent to patients automatically with a level of advice or guidance that patients can be satisfied.  

This research project has evolved into five phases. Now in its third year, it has reached the fourth phase of HDEC-approved early-phase clinical trials with patients at a small number of practices.  

Phase one: 

Phase one explored the feasibility and clinical considerations of laboratory results, including interviewing patients, primary and secondary care health practitioners, and medical laboratory providers. The findings of this stage informed that both patient expectations and provider processes need to be managed to be successful. Ethics advice was sought and was initially given out-of-scope determination.

Phase two: 

Phase two was developing a technology demonstrator proof of concept that would have supported capturing the clinical reasoning logic and patient-provider behaviour requirements outside of clinical records in the practice management system. However, during the development of this phase, it was clearly identified that there was a need to increase governance to ensure clinical and consumer viability. Following the demonstration of proof of concept, it was deemed that more robust mechanisms were required, which resulted in an Independent Advisory Panel being established to oversee developments.

Phase three: 

Phase three involved resetting expectations by robustly redeveloping research protocols that required full HDEC ethics approval. Final approval was given after new patient-orientated recruitment and consenting requirements were needed before any data can be collected.

Phase four: 

Phase four is the development and implementation of a real-world research system within Vensa’s clinical platform with early-stage clinical trials. Several changes have occurred since this development started as the realities of laboratory management at general practices required significant operational customisations. It is anticipated up to 800 patients may participate across up to three general practices in Auckland, in a number of rounds to measure the effectiveness of different feedback advice,  understanding patterns of reasons for laboratory ordering and the impact on risks, and clinical reasoning logic in interpreting patient results.

Phase five: 

Phase five is expected to begin in 2022, with a pragmatic implementation-research clinical trial with a dozen practices across New Zealand. This is intended to increase the volumes of data points to ensure clinical patterns can be stratified for decision support use.