A Single-Technology Assessment
FreeStyle Libre Flash Glucose Self-Monitoring System: A Single-Technology Assessment
Health technology assessment
|Published
This assessment will focus on FreeStyle Libre, flash glucose monitor for insulin treated individuals with type 1 and 2 diabetes (“Type 1 and 2 DM”).
Summary
Background
Diabetes mellitus (DM) has become one of the most common public health problems world-wide. According to the 2014 Norwegian Public Health report, diabetes affects an estimated 4.3% of the Norwegian population. Diabetes is a metabolic disorder resulting from a defect in insulin production, secretion, action, or all. Type 1 and 2 are the two main types, with the prevalence of type 2 accounting for the majority (>85%) of diabetes. This assessment will focus on FreeStyle Libre, flash glucose monitor for insulin treated individuals with type 1 and 2 diabetes (“Type 1 and 2 DM”).
To achieve proper quality of life and reduce long-term problems, people are increasingly encouraged to take an active role in the management of their condition. Adequate treatment management, aimed at tight control of blood glucose, reduces the risk of the long-term complications of diabetes such as retinopathy, nephropathy, neuropathy, coronary heart disease, ischaemic stroke and peripheral vascular disease. ‘Management’ of the disease should be understood as a package including testing of blood glucose, taking insulin (i.e., multiple daily insulin injections, using an insulin pump), using anti hyperglycemic drugs, or adopting lifestyle interventions such as diet and physical activity.
In recent years, and available in Europe since 2014, the FreeStyle Libre System - a ‘wireless’ method using a sensor for monitoring interstitial fluid glucose - was introduced to help individuals with type 1 and 2 DM achieve better glucose control. The system, unlike others, does not require finger prick calibration, since that functionality is embedded into the core technology. Also, unlike other systems, the individual has to take active action to get access to the real time glucose value, by leading the receiver over the sensor. Similarly to other continuous glucose monitoring options, it relies on the individual to take action on the information retrieved.
Objective
Our goal was to assess the clinical effectiveness, cost effectiveness and safety of FreeStyle Libre for individuals with type 1 and 2 DM.
Methods
We conducted a systematic review according to standard methods to summarise the evidence. The study populations were insulin treated individuals with Type 1 or 2 DM, the intervention was FreeStyle Libre, and the outcomes were HbA1c, hypo and hyperglycaemia, quality of life, patient satisfaction, pain, and adverse events.
We searched databases, trial registries, health technology assessment agencies websites and grey literature from inception to January 2017 with no language restrictions. Two reviewers independently screened the titles and abstracts of all records identified by searches, discussed any discrepancies and solved them by consensus. We obtained full text copies of all studies deemed potentially relevant and the same two reviewers independently assessed these for inclusion; solving any disagreements by consensus. One reviewer extracted data relating to study details, participants, intervention, and comparator, using a piloted, standard data extraction form. A second reviewer checked data extraction and any disagreements we resolved by consensus. The assessment of the methodological quality of each included study was based on the Cochrane Collaboration risk of bias tool. Quality assessment of evidence was carried out independently by two reviewers. We solved any disagreements by consensus. Meta-analysis was considered a suitable analysis for the data identified, despite heterogeneity. For some outcomes we employed a narrative synthesis.
Assessment of cost effectiveness
We assessed the cost-effectiveness estimates provided by the submitter of FreeStyle Libre compared to self-monitoring blood glucose (SMBG) for individuals with type 1 and 2 DM. The submitter used a commercially available cost-effectiveness model, IMS CORE diabetes model (IMS CDM) for this assessment. The model is internet based, with a Markov application, for individuals >18 years. The interactive simulation predicts the long-term health outcomes and costs associated with the management of type 1 and 2 DM. The model consist of 17 sub-models designed to simulate diabetes related complications, nonspecific mortality, and costs over time. As the model simulates individual patients over time, it updates risk factors and complications to account for disease progression. However, this model received from the submitter, lacks transparency, and made it difficult to gain a firm understanding of the factors that determine how patients progress through the model, assumptions and parameters effect outcomes and to assess the validity of the model. Because the Norwegian Institute of Public Health did not have complete access to the model, it was not possible to perform a full assessment of the model or to modify underlying assumptions and parameters in order to independently assess the impact on reported results. Furthermore, the documentation package did not include any sensitivity analysis, which is essential for considering the validity and robustness of results from economic evaluations.
Results
We included two randomized controlled trials (RCTs) in the review. These studies compared FreeStyle Libre to SMBG. Also, we found several publications investigating the accuracy of the device, however, the study designs of these studies (single arm) did not meet the inclusion criteria of this evaluation and, although we compiled them for information, they were excluded from the synthesis. The information derived from these single arm studies are potentially important to validate the sensitivity and specificity estimates of FreeStyle Libre. In addition, we found other European assessments conducted in the past 6 to 8 months. The included RCTs reported data on middle aged adults from European countries with type 1 and 2 DM at 6 months post intervention. We rated the studies’ risk of bias as unclear to high risk.
Main findings from these trials are that FreeStyle Libre may slightly improve treatment satisfaction, time spent with glucose in range 3.9 to 10 mmol/L, number of nocturnal events with glucose levels <3.1 mmol/L within 7h, and time spent with glucose levels >13.0 mmol/L in comparison to SMBG. FreeStyle Libre lead to little or no difference in quality of life and HbA1c level in comparison to SMBG. The evidence is uncertain about whether FreeStyle Libre leads to an improvement in time and events with glucose <3.9mmol/L within 24 h, time with glucose <3.1 mmol/L at night within 7 hours, and time with glucose > 10 mmol/L.
The submitted economic model runs a 40-year time horizon. The submitter´s basecase suggested that the technology is dominant for individuals with type 1 DM, i.e. that FreeStyle Libre is a cheaper and more effective technology. According to submitter´s base case, individuals with type 2 DM the incremental cost-effectiveness ratio (ICER) was calculated to be NOK 235,673 per QALY (whole study population) and NOK 243,434 per QALY (under 65 years). As the model received by the submitter was neither sufficiently transparent nor sufficiently flexible to allow changes, we have not been able to produce alternative incremental cost-effectiveness ratios (ICERs). From a healthcare perspective, the submitter has calculated a budget impact for type 1 DM to have a total added cost the fifth year after adoption of the technology. Further, the submitter calculated a budget impact for type 1 and 2 DM that lead to a cost saving on the fifth year after adoption of the technology. The submitter did not calculate a budget impact for type 2 DM only.
We estimated that, from a healthcare perspective, the annual costs five years after introduction would be NOK 186 million added cost and NOK 91,7 million saved cost for type 1 and 2 DM alone, respectively, and NOK 94 million added cost for type 1 and 2 DM combined.
Conclusions
Overall, the evidence for the intervention of interest was limited but suggests that FreeStyle Libre increases treatment satisfaction, reduces some hypo- and hyperglycaemic measures (increases time with glucose in range 3.9 to 10 mmol/L, reduces time and number of events with glucose <3.9 in 24 hours, number of glucose <3.1 night events and time with glucose >13 mmol/L) and has similar serious adverse events than SMBG, without differences in other outcomes including HbA1c and quality of life.
The quality of the included studies was generally low and there were only two small studies including middle aged adults.
Several inconsistencies lead us to question the result of the submitted health economic report. Specifically, the submitted model included several input data that did not match the input data described in the submitted documentation package, and nor did it match the input data found in other literature.
The most challenging issue is that the model is not sufficiently transparent or flexible, since we did not have access to the complete model. Therefore, we were not able to assess how the possible adjustments would affect the results provided by the submitters.
Suggested research priorities
- Independent research for FreeStyle Libre will be important
- Diabetes affects the life of children, adolescents and their caregivers in many ways, as well as pregnant women. Independent research including these groups is warranted
- The clinical effectiveness of FreeStyle Libre needs to be investigated in different conditions, for example, among individuals with poor self-monitoring adherence, newly diagnosed, impaired awareness of hypoglycaemia, and in addition to training and education components
- FreeStyle Libre compared to other continuous monitoring systems is warranted
- Pain is a major determinant of diabetes treatment adherence, especially for children, and it should be included as an individual outcome in future trials
- Future trials should include longer term follow up and quality of life outcome assessments at various points to inform improved clinical and cost effectiveness modelling
