FormalPara Key Summary Points

Why carry out this study?

Therapeutic inertia in type 2 diabetes, defined as a failure to intensify treatment despite poor glycemic control, can lead to an increased incidence of diabetes-related complications and associated costs that could be avoided with timely interventions.

The present study aimed to evaluate the health and economic burden of therapeutic inertia in people with type 2 diabetes in Saudi Arabia.

What was learned from the study?

Achieving glycemic control immediately was associated with improved or equal life expectancy and quality-adjusted life expectancy and cost savings in all scenarios compared with delays in achieving glycemic targets.

Therapeutic inertia was associated with a substantial health and economic burden in Saudi Arabia, with interventions and initiatives that can help to reduce therapeutic inertia likely to improve health outcomes and reduce healthcare expenditure.

Introduction

Type 2 diabetes is a chronic metabolic disease characterized by elevated blood glucose levels, affecting more than 537 million adults worldwide in 2021 [1]. In Saudi Arabia, diabetes has a particularly high prevalence, affecting more than 17.5% of the population (compared with a prevalence of 9.8% globally), and was associated with USD 7.5 billion of healthcare expenditure in 2021 [1, 2]. Landmark studies, such as the United Kingdom Prospective Diabetes Study (UKPDS), the Action to Control Cardiovascular Risk in Diabetes (ACCORD) study, and the Action in Diabetes and Vascular Disease-PreterAx and DiamicroN Controlled Evaluation (ADVANCE) study, demonstrated that short-term reductions in glycated hemoglobin (HbA1c) led to a reduced incidence of long-term diabetes-related complications, which can improve individuals’ quality of life while reducing costs for the healthcare payer [3,4,5,6,7]. This was demonstrated by a real-world study in Saudi Arabia that associated higher HbA1c levels with increased costs, with HbA1c categories of < 7.0%, 7.0–9.0%, and > 9.0% associated with direct medical costs of USD 1384, USD 2036, and USD 3105, respectively [8]. Moreover, a longitudinal retrospective cohort study in Saudi Arabia indicated that 68–80% of people with type 2 diabetes developed one cardiovascular complication, and 19–31% developed multiple cardiovascular complications (comprising angina, atrial fibrillation, coronary artery disease, heart failure, myocardial infarction and stroke), across a 3-year follow-up period [9]. These findings emphasize the need for early therapies and initiatives which can lower individuals’ blood glucose levels and subsequent long-term complications (both macrovascular and microvascular). Published treatment guidelines therefore often focus on controlling HbA1c as a key surrogate marker for people with type 2 diabetes [10, 11].

Within the last decade, modern treatments for type 2 diabetes, including sodium-glucose cotransporter-2 (SGLT-2) inhibitors and glucagon-like peptide-1 (GLP-1) receptor agonists, have become commercially available and are associated with improved efficacy and low risks of hypoglycemia compared to older classes of antidiabetics [10]. However, a number of people with type 2 diabetes often experience therapeutic inertia, defined as a failure to intensify treatment in a timely manner when needed (e.g., when experiencing poor glycemic control) [12]. Therapeutic inertia has been identified as a key issue in Saudi Arabia, with a country-specific study highlighting that only 15% of physicians prescribed GLP-1 receptor agonists for people with type 2 diabetes at the correct time [13, 14]. Moreover, data from a cross-sectional self-administered survey in Saudi Arabia have shown that SGLT-2 inhibitors and GLP-1 receptor agonists have low use rates when first-line therapy with other medications is insufficient (12% and 5%, respectively), and this remains low even in populations with established cardiovascular disease (31% and 15%, respectively) [15]. Studies in Saudi Arabia and the UK have also highlighted a particularly high level of therapeutic inertia when initiating injectable therapies, with common barriers to initiation including fear of injection, lack of education in people with type 2 diabetes, fear of hypoglycemia, and difficult administration [13, 14, 16, 17]. Physicians in Saudi Arabia have indicated reluctance to initiate insulin due to factors relating to people with type 2 diabetes, including expected non-adherence to blood glucose measurements, appointments or prescribed therapies, and individual refusal [16]. Reducing therapeutic inertia by addressing these concerns could therefore improve health outcomes while reducing overall costs for the healthcare payer in Saudi Arabia.

Previous analyses performed in the US, UK, and Sweden have demonstrated that short-term reductions in HbA1c can have a substantial impact on life expectancy and costs in country-specific populations with newly diagnosed diabetes [18,19,20]. Given the epidemiological burden of type 2 diabetes in Saudi Arabia and country-specific studies demonstrating a high level of therapeutic inertia, the aim of the present analysis was to evaluate the health and economic burden associated with therapeutic inertia and delayed time in achieving HbA1c targets in a Saudi-specific cohort with inadequate glycemic control on first- or second-line therapy.

Methods

Choice of Model and Approach

The IQVIA Core Diabetes Model (v.9.0; IQVIA, Basel, Switzerland) was used to project health and economic outcomes. Previous publications have described the structure, functions, assumptions, and capabilities of the model in detail, as well as performing two model validations [21,22,23]. In short, the model is a non-product-specific diabetes policy analysis tool comprised of several sub-models, each with a semi-Markov structure, to simulate diabetes-related complications and mortality. Patient-level simulations are performed based on patient risk factors, treatment approaches for microvascular and end-stage complications, and frequency of screening for microvascular complications. Model outcomes include life expectancy and quality-adjusted life expectancy, mean time to onset and cumulative incidence of diabetes-related complications, costs, and incremental cost-effectiveness ratios.

Outcomes were projected over a variety of time horizons (see next section), discounted at 3.0% per annum (in line with guidance for the Saudi Arabian setting), and projected using the UKPDS 68 risk equations [24, 25]. Background mortality was captured from Saudi Arabian-specific life tables published by the World Health Organization [26]. All analyses were performed with a first-order Monte Carlo simulation approach.

Modeled Scenarios and Parameter Progression

Scenarios were selected to reflect the different levels of poor glycemic control and therapeutic inertia in Saudi Arabia, with additional variation in the time horizon applied to capture the different interests of healthcare payers. Variations were performed in three areas: three levels of baseline HbA1c (8.0%, 9.0%, and 10.0%); five levels of therapeutic inertia (delays in achieving target HbA1c of 1, 2, 3, 4, and 5 years); and six time horizons (3, 5, 7, 10, 15, and 50 years). Target HbA1c was defined as 7.0%, based on clinical practice guidelines for the treatment of diabetes in Saudi Arabia [11]. In each analysis, individuals were modeled to achieve target HbA1c in the first year of the analysis (defined as ‘immediate control’) or after a 1- to 5-year delay (defined as ‘poor control’), with scenarios designed to be hypothetical rather than reflecting any specific interventions. HbA1c was assumed to remain at each of the defined levels throughout all analyses (that is, no parameter progression equations were applied). Other physiological parameters, including blood pressure, serum lipid levels, and body mass index (BMI) were also assumed to remain constant at baseline throughout each analysis. No hypoglycemic event rates were applied.

Baseline Cohort Characteristics

Baseline cohort characteristics were sourced from a Saudi-specific, 5-year longitudinal cohort review, representing people with type 2 diabetes with inadequate glycemic control on first- or second-line therapy (Table 1) [27]. All inputs for an analysis in the IQVIA Core Diabetes Model version 9.0 were captured in the dataset, with the exception of duration of diabetes and history of neuropathy, which were assumed to be zero (Supplementary Material, Table S1). A cohort of 2226 adult patients with type 2 diabetes were originally captured, with data extracted for a subpopulation of 638 individuals with an available baseline HbA1c ≥ 6.5% and receiving at least one treatment with a non-missing start date. First- and second-line therapy captured a range of medications, including metformin (82% of people), sulfonylureas (52%), thiazolidinediones (0.3%), dipeptidyl peptidase-4 (DPP-4) inhibitors (44%), SGLT-2 inhibitors (3%), GLP-1 receptor agonists (4%), and insulin therapies (1–20%).

Table 1 Baseline cohort characteristics applied in the analyses
Full size table

Costs and Utilities

All analyses were performed from a societal perspective, capturing direct costs of treating diabetes-related complications informed by published sources (inflated where necessary to 2023 Saudi Riyals [SAR]) and indirect costs associated with lost workplace productivity calculated via a human capital approach based on Saudi-specific salaries and days off work estimates published by Sørensen and Ploug and Persson et al. (Supplementary Material, Tables S2 and S3) [28, 29]. No acquisition costs relating to antidiabetic medications were included in the analyses, as the aim was to assess the clinical and economic burden of different levels of glycemic control, rather than specific interventions for type 2 diabetes. Health-state utilities and event-based disutilities relating to quality of life were sourced from a 2014 systematic review by Beaudet et al., which informs the default utility set in the IQVIA Core Diabetes Model (Supplementary Material, Table S4) [30]. All results were reported in SAR and in euros (EUR), using a conversion rate of SAR 1 = EUR 0.247.

Compliance with Ethics Guidelines

This article is based on previously conducted studies and does not contain any studies with human participants or animals performed by any of the authors.

Results

Cost Outcomes

Modeling projections showed that failure to intensify therapy in a timely manner had a substantial impact on the economic burden associated with type 2 diabetes in Saudi Arabia, with immediate glycemic control associated with cost savings in all analyses versus remaining in poor glycemic control (Table 2; Fig. 1). These arose due to avoidance of diabetes-related complications, most notably cardiovascular complications versus shorter delays over shorter time horizons and renal complications versus longer delays over longer time horizons. For individuals with a baseline HbA1c of 8.0%, combined (direct and indirect) cost savings ranged from SAR 411 (EUR 102) per person versus a 1-year delay over a 3-year time horizon, to SAR 4366 (EUR 1078) per person versus a 5-year delay over a 50-year time horizon. Increasing baseline HbA1c to 9.0% resulted in increased cost savings with immediate glycemic control, with projections ranging from SAR 844 (EUR 208) per person versus a 1-year delay over a 3-year time horizon to SAR 11,754 (EUR 2903) per person versus a 5-year delay over a 50-year time horizon. The greatest cost savings were observed with a baseline HbA1c of 10.0%, with immediate glycemic control associated with savings ranging from SAR 1315 (EUR 325) per person versus a 1-year delay over a 3-year time horizon to SAR 21,422 (EUR 5291) per person versus a 5-year delay over a 50-year time horizon. Cost savings were observed in all scenarios, but were more pronounced at higher baseline HbA1c levels, versus longer delays in achieving target HbA1c, and over longer time horizons (Table 2; Fig. 1).

Table 2 Per-person cost saving values projected in the analyses
Full size table
Fig. 1
figure 1

Per-person cost savings projected in the analyses. SAR 2023 Saudi Riyals

Full size image

Life Expectancy Outcomes

Short- and long-term projections indicated that therapeutic inertia was associated with a considerable impact on health outcomes in Saudi Arabia. Immediate glycemic control was associated with equal life expectancy in 4 out of 75 scenarios and improved life expectancy in 71 out of 75 scenarios versus remaining in poor glycemic control (Fig. 2). Scenarios in which equal life expectancy (measured to two decimal places) was observed were those with a baseline HbA1c of 8.0% versus a delay in achieving target HbA1c of 1 or 2 years over 3- and 5-year time horizons, or a baseline HbA1c of 9.0% versus a delay in achieving target HbA1c of 1 year over a 3-year time horizon. In all other analyses, life expectancy was improved with immediate glycemic control versus delays in achieving target HbA1c, with benefits ranging from an additional 0.01–0.41 years of life gained per person. More pronounced improvements were associated with higher HbA1c levels at baseline and observed versus longer delays in achieving target HbA1c over longer time horizons (Fig. 2).

Fig. 2
figure 2

Incremental per-person life expectancy projected in the analyses

Full size image

Quality-of-Life Outcomes

Quality-adjusted life expectancy was projected to be markedly improved with immediate glycemic control versus a delay in achieving target HbA1c in Saudi Arabia (Fig. 3). Equal quality-adjusted life expectancy (measured to two decimal places) was observed in one analysis (baseline HbA1c of 8.0% vs. a 1-year delay over a 3-year time horizon). In all other analyses, immediate glycemic control was associated with improvements in quality-adjusted life expectancy versus remaining in poor glycemic control. Per-person benefits ranged from an additional 0.01–0.46 quality-adjusted life years (QALYs) gained, with improvements correlated with higher HbA1c levels at baseline, longer delays in achieving target HbA1c in the comparator arm, and longer time horizons (Fig. 3). For example, for people with a baseline HbA1c of 8.0%, improvements in quality-adjusted life expectancy of 0.03 QALYs were observed versus a 3-year delay over a 10-year time horizon, compared with benefits of 0.11 QALYs observed versus a 5-year delay over a 50-year time horizon. Similarly, applying a higher baseline HbA1c of 9.0% led to improvements ranging from 0.01 QALYs versus a 1-year delay over a 3-year time horizon to 0.28 QALYs versus a 5-year delay over a 50-year time horizon.

Fig. 3
figure 3

Incremental per-person quality-adjusted life expectancy projected in the analyses. QALY quality-adjusted life years

Full size image

Discussion

The present study is among the first to evaluate the burden of poor glycemic control due to therapeutic inertia in Saudi Arabia. While similar results have been reported for the US, UK, and Sweden, these country-specific estimates should provide important information for healthcare payers in Saudi Arabia [18,19,20]. These results highlight the need for therapies that can control blood glucose levels, while providing attractive administration options for people with type 2 diabetes, to avoid potential therapeutic inertia and increased levels of future healthcare expenditure due to an increased incidence of diabetes-related complications [3,4,5,6,7,8]. These complications are also associated with a substantial quality-of-life burden that can translate to considerable benefits over individual lifetimes, and the present analysis is among the first to report these outcomes. These results also emphasize the importance of modeling long-term outcomes associated with diabetes to fully capture the incidence of diabetes-related complications and the impact of improved glycemic control. The scope for improvement in Saudi Arabia is highlighted by the low proportions of people receiving modern antidiabetic therapies at baseline in the extracted cohort, with only 3% receiving SGLT-2 inhibitors and only 4% receiving GLP-1 receptor agonists as second-line therapies (see Section "Baseline cohort characteristics"), and recent prescribing patterns, which indicate a low use of SGLT-2 inhibitors (12%) and GLP-1 receptor agonists (5%) as second-line therapies [15]. As well as improvements in glycemic control, these medication classes have been associated with both cardioprotective and renoprotective effects, which could lead to further benefits not captured in the present study [10, 31, 32]. Use of efficacious, orally-administered medications such as these can therefore have a two-fold impact by improving health outcomes and quality of life while potentially reducing costs for the healthcare payer.

Therapeutic inertia in type 2 diabetes, defined as a failure to intensify therapy despite poor glycemic control, has been shown to exist at all stages of the diabetes treatment algorithm, but individuals’ aversion to once-daily injectable therapies is particularly pronounced [33,34,35]. In Saudi Arabia, this has been recently demonstrated in studies evaluating once-weekly injections versus once-daily injections, and a once-daily oral GLP-1 receptor agonist versus a once-weekly formulation [13, 14, 16]. Published data have also demonstrated a notable quality-of-life benefit associated with once-daily oral administration compared with once-weekly and once-daily injectable therapies, as well as once-weekly versus once-daily injectables [36, 37]. Poor treatment adherence also remains a challenge in diabetes management, with a reported 50% of people suffering from suboptimal adherence [10, 16, 38]. Interventions and initiatives that can improve adherence, including education of physicians and people with type 2 diabetes around the efficacy and safety of available medications with novel administration methods, could thereby reduce the burden associated with therapeutic inertia. Indeed, improved communication between physicians and patients can also help to relive key apprehensions around treatment intensification [39]. Efforts to improve diabetes management in Saudi Arabia between 2018 and 2021 have shown positive results, but further improvements are still possible [40].

The present analysis captured a relatively experienced cohort compared with previous modeling studies evaluating the burden of poor glycemic control due to clinical inertia, with the majority of people receiving more than one antidiabetic medication [18,19,20] That acknowledged, initiatives that diagnose and treat people with type 2 diabetes at the earliest possible opportunity are likely to increase the levels of cost savings and clinical benefits projected in the present study, with earlier and tighter glycemic control likely to lead to a further reduced incidence of diabetes-related complications over the long term. Indeed, the variations in delays in achieving glycemic control in the present analyses were in line with those reported in routine clinical practice for type 2 diabetes and intended to provide a range of scenarios to clearly demonstrate the importance of immediate glycemic control [33, 34, 41, 42]. Moreover, variations in the time horizons of the analyses were intended to show that, despite long-term diabetes-related complications forming the largest proportion of associated expenditure, cost savings and notable clinical benefits for people with type 2 diabetes can be achieved over the short term should glycemic control be realized and maintained (Figs. 1, 2, 3). Comparison between the estimated life expectancy for the general population in Saudi Arabia (74.3 years from birth) and the modeled life expectancy in the immediate glycemic control arm of the present study (73.2 years from a baseline age of 49.1 years) also demonstrates that a near-normal life expectancy is possible in people with type 2 diabetes should glycemic targets be achieved [43]. The longitudinal cohort study that informed the baseline cohort characteristics of the present analysis were also used to demonstrate a substantial burden of illness for type 2 diabetes in Saudi Arabia [27]

Limitations of the analysis included the inclusion of only changes in HbA1c. Treatment effects capturing changes in systolic blood pressure, serum lipid levels, or BMI were not applied, and no lifestyle interventions such as smoking cessation or increased physical exercise were modeled. Given that modern treatments of type 2 diabetes have multifactorial benefits beyond reductions in HbA1c (with SGLT-2 inhibitors and GLP-1 receptor agonists associated with significant weight loss benefits, low risks of hypoglycemia, and cardiorenal protective effects), it is possible that the projected cost savings and benefits in quality of life were underestimated [10]. Moreover, alongside availability of multifactorial therapies, diabetes management is becoming increasingly holistic, with individualized treatment regimens and targets often recommended [10]. However, the aim of the study was to estimate the clinical and economic burden associated with poor glycemic control due to therapeutic inertia, and the chosen approach of maintaining all physiological parameters bar HbA1c as constant with a Saudi-specific HbA1c target was considered the fairest way of answering this research question [11]. Treatment costs were also excluded from the analysis for this reason, as there are a variety of therapy options to improve glycemic control and the present study identifies the budget available without increasing overall healthcare expenditure. Any health economic analysis evaluating specific medications should balance the benefits gained against the acquisition costs and adverse events of the interventions, and these aspects should be fully considered in a product-specific cost-effectiveness analysis.

Conclusion

In projections over both the short and long terms, immediately bringing HbA1c to target levels was associated with improved clinical outcomes and a reduced economic burden compared with prolonged periods spent in poor glycemic control due to therapeutic inertia in Saudi Arabia. Interventions and initiatives that can reduce therapeutic inertia and achieve improved glycemic control in the country should provide crucial benefits for people with type 2 diabetes and cost savings for healthcare payers.