Post-market Review Products Used in the Management of Diabetes Report to Government Stage 2: Insulin Pumps February 2015

Executive Summary

Background and context

This Report contains the findings of Stage 2 of the Post-market Review of Products used in the Management of Diabetes focussing on insulin pumps (the Insulin Pumps Review). The objective of the Insulin Pumps Review, in line with the Diabetes Review Terms of Reference (ToR 8-10), was to evaluate the clinical outcomes and cost-effectiveness of insulin pump therapy for people with type 1 diabetes, and the clinical and financial eligibility criteria for the Insulin Pump Programme. The Insulin Pumps Review was conducted according to the Postmarket Review Framework (as at February 2014).

The Type 1 Diabetes Insulin Pump Programme (the Programme) aims to increase the affordability of insulin pump therapy for children up to 18 years of age from low-income families without private health insurance by subsidising the cost of an insulin pump. The Programme also provides funding for part of the cost of pump consumables, including infusions sets, cannulas, and tubing. The Juvenile Diabetes Research Foundation (JDRF) administers insulin pumps under the Programme, while consumables are supplied through the National Diabetes Services Scheme (NDSS) administered by Diabetes Australia. Programme funding totals approximately $7.1 million over the four years from 201213 to 201516.

There are approximately 118,600 Australians with type 1 diabetes (ABS 2013), of whom approximately 10,510 are currently using an insulin pump (AIHW 2012b). As at February 2014, 466 pumps were supplied under the Programme. Families accessing a subsidised insulin pump under the Programme are able to choose different brands of pumps. Subsidised pumps are generally those on the Department of Health Prostheses List, which is designed for private health insurance purposes and is managed by the Department. As continuous glucose monitoring sensors and transmitters are not currently subsidised under the Insulin Pump Programme or by private health insurers, sensor-augmented pump therapy was considered outside of the scope of the Review.

Review findings

ToR 8 focussed on clinical outcomes of insulin pump therapy and the Review considered evidence from randomised comparative trials and observational studies. The systematic literature review of randomised controlled trial (RCT) evidence showed that this evidence does not conclusively support the superior efficacy or safety of insulin pump therapy over multiple daily injections in any age group. The overall strength of this evidence for decision-making was low to very low, with minimal benefit from insulin pump therapy on HbA_1c and no effect on hypoglycaemia. The observed benefit of insulin pump therapy compared to multiple daily injections showed a reduction in HbA_1c levels, ranging from 0.1–0.22% in children and adolescents, and 0.19–0.30% for adults. The difference did not reach the 0.5% reduction generally accepted to be of clinical significance (Clar 2010; Cummins 2010). However, there is no consensus on the best minimum clinical difference and a smaller reduction may be considered important from a public health perspective if achieved by most people using pumps in place of multiple daily injections (Farmer 2012). No trials were identified that assessed longterm health outcomes.

The systematic review of observational evidence indicated that children, adolescents, and adults appear to achieve greater reductions in HbA_1c and have a lower risk of severe hypoglycaemia with insulin pump therapy compared to multiple daily injections. Some long term studies showed statistically significant reductions in HbA_1c ranging from 0.40.7% in children and adolescents using insulin pump therapy at 2–5 years follow-up. A retrospective cohort study undertaken in children and adolescents in Australia showed a significant reduction in HbA_1c of 0.7% with insulin pump therapy over 5 years and reduced rates of severe hypoglycaemia and hospitalisation for ketoacidosis (Johnson 2013).

Before-and-after studies showed statistically significant reductions in HbA_1c of 0.4–2.6% in adults using insulin pump therapy at 2–6 years follow-up, with the majority showing reductions of 0.40.7%. The largest cohort study of adults showed a statistically significant reduction in HbA_1c with insulin pump therapy of 0.2% at five years (Carlsson 2013). For women with type 1 diabetes who are pregnant or planning a pregnancy, the majority of observational studies did not show a significant difference between insulin pump therapy and multiple daily injections on diabetes management, maternal pregnancy outcomes or newborn outcomes. There were similar results from systematic reviews of trial evidence (Farrar 2007; Mukhopadhyay 2007).

The observational studies lacked randomisation of the choice of selected treatment resulting in a high risk of selection and publication bias. Therefore, results should be interpreted cautiously. Despite the limitations of observational studies, the Review considered these studies as the patients included may be more reflective of the patients currently using insulin pump therapy in Australia, such as those with diabetes complications or hypoglycaemia unawareness, who were generally excluded from RCTs.

A review of cost-effectiveness studies of insulin pump therapy was undertaken consistent with ToR 9. Programme costs, utilisation, and affordability and access issues were also considered. The literature review did not identify any studies on the comparative effectiveness of different insulin pump brands.

On average, insulin pump consumables (excluding insulin) cost consumers $310 per annum. The average cost to Government for each insulin pump supplied through the Programme is $14,779 over four years (the warranty period), or $3,695 per annum. This estimate is based on the median pump subsidy amount ($6,400) and the cost of consumables for four years ($2,098 per annum).

Four cost-effectiveness studies in Australia, Canada, UK and US comparing insulin pumps to multiple daily injections were identified. All studies used the CORE model (Palmer 2004). The cost-effectiveness studies, while assuming a large reduction in HbA_1c with insulin pump therapy of 0.9% in one study and 1.2% in three studies, suggest a cost per quality-adjusted life year (QALY) gained of AUD$28,874 to AUD$96,220 at 2013 prices, with the UK study providing the upper estimate.

Sensitivity analyses showed that the incremental cost-effectiveness ratios (ICERs) were most sensitive to the estimated reductions in HbA_1c. Assuming reductions in HbA_1c of 0.51%–0.675% with insulin pump therapy compared to injections, the ICERs per QALY ranged from AUD$63,274 in the US study to AUD$292,952 in the UK study (converted to 2013 prices). The US, Canadian and Australian studies assumed a pump life span of seven or eight years, while the standard warranty length for an insulin pump is four years. These studies may have overestimated the cost-effectiveness of insulin pump therapy compared to multiple daily injections if the average pump life span is less than that assumed. In the US study, reducing the pump life span to four years increased the ICER by 54%. The ICERs were sensitive to changes in severe hypoglycaemic events only when the cost of severe hypoglycaemia was high (2007 USD$1,234). The studies did not consider improvements in quality of life with the exception of fear of hypoglycaemia. Stakeholders considered quality of life to be a key benefit of insulin pump therapy.

The investigation of the Programme’s usage and financial eligibility criteria highlighted several issues with affordability for the eligible population. JDRF made agreements in 2011 with some manufacturers to cover the cost of the patient co-payment for families on the lowest eligible incomes of under $71,230 (2013-14) per annum. This aided Programme access and led to the development of a waiting list in 2012-13, which was addressed by additional funding. The Prostheses List insulin pump prices increased on 1 August 2012, which subsequently increased the patient copayment, as the maximum subsidy is capped. Families on the lowest eligible incomes are now less likely to be able to afford the copayment without further subsidy or continued assistance from insulin pump manufacturers. For many other eligible families, private hospital insurance may be a more affordable option than to access the Programme. Median family income data indicates that families on eligible incomes between $71,230-$101,653 (2013-14) rarely access the Programme (JDRF data).

ToR 10 examined the clinical eligibility of the Programme, as well as a comparison of Australian and international guidelines for insulin pump therapy, and subsidy arrangements in other countries. This showed that most guidelines recommend that the decision to initiate insulin pump therapy should be at the discretion of a healthcare professional in consultation with the patient or carer. The literature shows that there is limited evidence regarding sub-groups that receive greater benefit from insulin pump therapy. This makes it difficult to define clinical criteria to ensure that the Programme is available to patients for whom insulin pumps are most likely to be effective and costeffective.

The Australasian Paediatric Endocrine Group (APEG) advised that the current Programme eligibility criteria are appropriate for children, and that there is not enough evidence on which to base any additional criteria. Australian and international clinical guidelines often recommend insulin pump therapy for: children under 12 years where multiple daily injections are impractical; women who are pregnant or trying to conceive; and people with poor glycaemic control. In the healthcare systems examined, most provided full or partial subsidy arrangements for an insulin pump, with a focus on providing access for children and young adults.

Stakeholder input

Stakeholders generally consider that there are clinical (effectiveness and safety) and quality of life benefits to insulin pump therapy. The main quality of life benefits associated with insulin pump use raised by stakeholders were: greater flexibility with meals and sleeping times; enhanced participation in social, school and sporting activities; and enhanced participation in employment, including for carers of children with type 1 diabetes.

Stakeholders were supportive of the continuation of the Programme and suggested expansion to pregnant women and those planning a pregnancy, and to adults with high initial HbA_1c or disabling hypoglycaemia (severe and/or unpredictable hypoglycaemia). Clinical groups acknowledged that the evidence does not conclusively support the clinical benefits of insulin pump therapy. However, they consider there is additional benefit from using insulin pumps rather than multiple daily injections in particular patient subgroups. With the exception of the age criterion, stakeholders generally agreed that the current Programme eligibility criteria were appropriate.

Stakeholders noted that success with either insulin pump therapy or multiple daily injections is dependent on the level of motivation and compliance of the patient/carer. They suggested that access to a range of insulin pumps allowed health professionals to tailor products to individual patient needs. Stakeholders considered that insulin pumps should be fully subsidised for those on the lowest eligible incomes under the Programme.