The objective of this review is to evaluate and summarize the evidence base for the effects of monetary intervention approaches (the use of positive monetary reinforcers and gains) on diabetes outcomes. A reproducible search using OVID Medline, PubMed, Scopus, and CINAHL was conducted. Articles published from database creation up to July 2024 were searched. Outcomes included hemoglobin A1c (HbA1c), LDL, BMI, blood pressure, quality of life (QOL), psychosocial factors, self-care behaviors, and diabetes complications. A total of 13 articles met inclusion criteria and were included for final synthesis. Looking at the monetary approach across each study, eight used financial incentives, three used a form of income supplementation, one used cash transfers, and one used a combination of income supplementation and financial incentives. Ten of the 13 studies found statistically significant and clinically meaningful changes in HbA1c. For participants receiving interventions, change in HbA1c ranged from 0.19% to 1.74% for interventions incorporating financial incentives, 0.7% to 1.3% for interventions incorporating income supplementation, and 0.2% to 0.7% for the study incorporating cash transfers. Overall, evidence supports the relationship between monetary approaches, diabetes-related outcomes, and self-care behaviors across monetary approaches. Future studies should consider comparison between different monetary approaches using designs that will allow identification of effective strategies. As these approaches are theoretically and structurally different, pathways identifying the underlying mechanisms of change are greatly needed to advance the field.
Burden of Diabetes
Diabetes affects over 800 million people worldwide (1) and is associated with heart disease, stroke, and kidney disease and contributes to functional limitations such as blindness and lower-limb amputation (2). Type 2 diabetes, accounting for 90–95% of all diabetes cases, exerts significant impact on the world economy (1–5). As a result, health agendas across the World Health Organization, Centers for Disease Control and Prevention, and American Diabetes Association continue to promote population-based efforts directed at decreasing the burden of disease for adults at risk for or living with diabetes (6–8). Despite this, recent estimates show only 11.1% of adults living with diabetes meet more stringent clinical recommendations for the so-called ABCs of diabetes (i.e., hemoglobin A1c [HbA1c] <7%; blood pressure <140/90 mmHg; non-HDL–cholesterol <130 mg/dL; and nonsmoking) (1,9), and only 36.8% meet less stringent clinical recommendations (HbA1c <8%; blood pressure <140/90 mmHg; non–HDL-cholesterol <160 mg/dL; and nonsmoking) (1,9). As such, there is an urgent need for population health approaches designed to reduce overall burden and optimize health long-term for people living with diabetes (5).
Evidence shows that the use of monetary approaches, as positive reinforcers and gains, are associated with improving diabetes clinical outcomes and may be important for optimizing health long-term (10–12). Monetary approaches include the use of financial incentives, income supplementation, cash transfers, and universal basic income (UBI). Each intervention approach uses monetary gains, versus penalties, as the premise for behavior change, yet each intervention is theoretically and structurally different. While each approach shows promise in reducing risk for diabetes complications by significantly lowering HbA1c, blood pressure, and cholesterol, little has been done to summarize the evidence for these intervention approaches. Given the similarities and differences in structure of interventions and associations on outcomes for diabetes, even less has been done to provide clear definitions and theoretical underpinnings of each.
Monetary Intervention Approaches: Terms and Definitions
Financial Incentives
Financial incentives are defined as a monetary benefit offered to encourage behavior or actions that otherwise would not take place. A financial incentive motivates actions that otherwise might not occur without the monetary benefit (13). The financial incentive schemes can vary widely depending on the behavioral targets (14); however, typical financial incentives range from a few U.S. dollars a month to US$125/month, are tied to specific behaviors and/or outcomes, are not guaranteed, and are not designed to supplement income. Financial incentives differ from financial disincentives, which serve as fines, penalties, or fees associated with not performing a desired behavior or ongoing behaviors (15).
Income Supplementation
Income supplementation is a monetary approach that is often conflated with other approaches and has no agreed-upon definition. For the purposes of this review, we examined the definitions of both income and supplementation to develop a working definition that will allow identification of interventions. Income is defined as money received, especially on a regular basis, for work or through investments (16). Supplementation is defined as the addition of an extra element or amount (16). Therefore, our working definition of income supplementation is the provision of minimal monetary support through cash, vouchers, and/or coupons for the purpose of supplementing earned income to support basic needs. Typical income supplementation ranges from US$100/month to US$250/month, are guaranteed, are designed to supplement income, and, while intended to be spent on basic needs, may not always be restricted to predefined categories of use.
Cash Transfer Programs
Cash transfer programs are defined as programs that provide assistance in the form of money to increase household income (17–20). Transfers may be given without the requirement that household members meet specified conditions (unconditional cash transfer [UCT]) or be contingent upon compliance with a specified set of conditions (conditional cash transfer [CCT]) (17–20). Typical cash transfer programs range from US$300/month to US$1,000/month, are guaranteed, and are designed to supplement income.
UBI
Universal basic income (UBI) is defined as financial support provided by a government or entity in the form of standard, recurring payments to all individuals without the need for prequalification, such as employment or means testing (16,21–22). Typical UBI programs range from US$1,000/month to US$1,500/month, but unlike cash transfer programs, they are not means tested and are available to all income levels.
Theoretical Underpinnings of Monetary Transfer Approaches: Economic Theory
Per economic theory, it is helpful to characterize monetary transfer intervention approaches (such as UBI, cash transfers, and income supplementation) according to three dimensions. First, whether it is universal or targeted to any specific group based on group characteristics. Second, whether it is unconditional or conditional on the individual satisfying prespecified compliance criteria. Third, whether the intervention represents a cash or an in-kind transfer (for example, the in-kind provision of housing or food) (23–24). According to these criteria, UBI is the least restrictive form of a transfer intervention, as it is characterized by universality, unconditionality, and a cash (rather than in-kind) transfer. Other forms of transfers, such as income supplementation, cash transfer programs, and broader financial incentives, all tend to be more restrictive in that they are often targeted to specific socioeconomic or demographic groups, are based on prespecified compliance criteria, and can take the form of both cash and in-kind transfers (e.g., in-kind food stamps provided through the U.S. Supplemental Nutrition Assistance Program) (23–24). From the perspective of welfare maximization, economic theory implies that people are better off (or at least as well off) under less restrictive intervention approaches; however, it is important to recognize that all noted transfer interventions (assuming individuals consume normal goods, the consumption of which increases with income) result in a positive income effect that increases individual welfare upon receipt (23–24). This provides a theoretical justification for the noted intervention approaches; however, it remains important to recognize that utility maximization may not coincide with the maximization of individual health or health care management and that psychological factors disregarded by traditional economic theory may influence the overall impact of monetary approaches, particularly those tied to financial incentive approaches.
Additional Theoretical Underpinnings of Financial Incentives: Behavioral Economic Theory
Behavioral economics is a well-established field that studies human decision-making by drawing on insights from both economics and psychology (14,24–28). The union of economic and psychological principles is a critical feature of the field, as traditional economic theory assumes rational and perfectly informed decision-making that is commonly found to be incongruent with observed patient behaviors (26). For example, while people living with diabetes may be aware that taking their medications as prescribed or following recommended diet and exercise regimes can help improve their long-term health, awareness and knowledge may not result in adoption and engagement. The behavioral economic explanation for this is that individuals’ costs of adopting behavior are commonly incurred long before the full benefits of their actions are realized and observed (26). The temporal difference between when the costs and benefits are incurred impact individuals’ behavior because people tend to discount the future more than the present and because of present bias, a phenomenon where short-term costs and benefits are given disproportionate psychological weight in comparison with future costs and benefits (29).
Per behavioral economic theory, financial incentives have a direct (positive) income effect (as previously noted) but also have an indirect psychological effect that is more ambiguous (30). The ambiguity of the psychological effect stems from it being possible for monetary incentives to also reduce desired behaviors, for example, by establishing reliance upon an external incentive without encouraging internal motivation (30). For most tasks, the direct (positive) income effect will tend to be larger than any negative psychological effect in the short run, so long as incentives are sufficiently large (30). This provides support for why financial incentives can help stimulate desired behavioral responses and changes within patients.
Current Gaps and Rationale for Scoping Review
As evidence continues to emerge and interventions using monetary approaches are developed, there is a need to delineate financial incentives, income supplementation, cash transfers, and UBI and to understand the evidence for how each affects diabetes outcomes in adults. Interventions that use monetary gains as an intervention component to improve outcomes are often multifaceted, making it difficult to isolate effective intervention approaches. Therefore, a scoping review was selected as a first step to understanding monetary approaches for improving diabetes clinical outcomes for people living with type 2 diabetes. Scoping reviews are a critical precondition to systematic reviews and meta-analyses, particularly for areas that span disciplines (31). In addition, scoping reviews are ideal for outlining the extant literature, unifying understanding for inconsistent use of terminology, and setting the stage for intervention development, health care practice, or health policy (31).
The purpose of this scoping review is to examine and summarize the evidence base for the effect of monetary intervention approaches, focusing on the differences between and effectiveness of financial incentives, income supplementation, cash transfers, and UBI interventions, on diabetes outcomes. This review will be specific to adults with type 2 diabetes, with outcomes focused on the ABCs of diabetes care, quality of life (QOL), and diabetes self-management behaviors, as outlined in guidelines set forth by the American Diabetes Association (3). Conclusions will focus on the state of the evidence and strategies associated with improved outcomes that can inform health care practice and health policy to improve outcomes for adults with type 2 diabetes. Consistent with scoping review methodology, this study will use a narrative synthesis where the heterogeneity and homogeneity across studies will be explored descriptively rather than statistically (31). This approach is recommended when multiple study designs are being evaluated or interventions are theoretically distinct and sets the stage for systematic reviews and meta-analyses.
Research Design and Methods
Information Sources, Search and Eligibility Criteria
Identifying, screening, and study selection were conducted using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) 2020 guidelines (32) (Fig. 1). Eligibility criteria for article selection were established by the authors a priori. A reproducible search strategy was used to identify evidence for the use of financial incentives, income supplementation, cash transfers, and UBI interventions to improve diabetes outcomes. Key databases were selected based on broad representation of the literature, including 1) Ovid MEDLINE, a robust database including international literature from over 5,600 journals across the fields of biomedicine, physical sciences, humanities, medicine, and health care, as well as allied fields; 2) PubMed, the National Center for Biotechnology Information (NCBI) database with >37 million citations spanning the biomedical sciences; 3) Scopus, representing over 300 disciplines and including literature from a global perspective; and 4) CINAHL, a collection of allied health and nursing literature representing >1,389 open access journals globally. Filters were applied to include articles that were published in English and that focused on adults. Articles published from database creation through July 2024 were searched. Medical Subject Heading (MeSH) terms and keywords representing financial incentives, income supplementation, cash transfers, and UBI and type 2 diabetes were used (Table 1). After duplicates were removed, articles were screened using eligibility criteria.
Search terms
| Monetary approach terms searched (title/abstract) combined with OR |
| Universal basic income, cash transfer, income supplement, income supplementation, financial incentive, monetary incentive, social welfare, social assistance, voucher, financial support, value voucher, food assistance, mobile money, e-cash, e-transfer, e-wallet, cash, coupon, direct payment, conditional transfer, conditional cash transfer, unconditional transfer, unconditional cash transfer, public assistance, food assistance, medical assistance, old age assistance, restricted transfer, unrestricted transfer |
| Diabetes search terms searched combined with OR |
| Diabetes (title/abstract/keyword), type 2 diabetes (title/abstract/keyword), diabetes mellitus, type 2 (MeSH), diabetes mellitus (MeSH) |
| Monetary approach terms searched (title/abstract) combined with OR |
| Universal basic income, cash transfer, income supplement, income supplementation, financial incentive, monetary incentive, social welfare, social assistance, voucher, financial support, value voucher, food assistance, mobile money, e-cash, e-transfer, e-wallet, cash, coupon, direct payment, conditional transfer, conditional cash transfer, unconditional transfer, unconditional cash transfer, public assistance, food assistance, medical assistance, old age assistance, restricted transfer, unrestricted transfer |
| Diabetes search terms searched combined with OR |
| Diabetes (title/abstract/keyword), type 2 diabetes (title/abstract/keyword), diabetes mellitus, type 2 (MeSH), diabetes mellitus (MeSH) |
Monetary approach term search and diabetes term search were combined with AND.
Eligible articles were 1) published in English, 2) incorporated any study design, including cross-sectional, cohort, randomized controlled trial (RCT), experimental, quasi-experimental, or pre-post study design, 3) focused on adults aged ≥18 years with type 2 diabetes, 4) evaluated the use of financial incentives, income supplementation, cash transfers, or UBI, and 5) included one or more of the following diabetes-related outcomes: HbA1c, LDL, BMI, blood pressure, QOL, self-efficacy, mortality, years of life lost, stress, and self-care behaviors. Clinical outcomes were selected based on clinical guidelines for diabetes care and standards of care measures set by the American Diabetes Association (3). This review was not restricted to the U.S. Exclusion criteria included monetary approaches used to change physician behavior, as the primary focus here is individuals with type 2 diabetes.
Study Selection, Data Extraction, and Data Quality
Study selection was based on an initial title and abstract review. Studies were evaluated for inclusion using a checklist that included eligibility criteria. Studies not meeting eligibility criteria were excluded. After the title and abstract review, full-text articles that met initial inclusion criteria were included for full-text synthesis. Initial and full-text review of studies were done separately and finalized by the authors. The checklist ensured consistent decision-making processes were followed for each article reviewed. After full-text synthesis, articles not meeting inclusion criteria were excluded for various reasons. Studies including populations with type 2 diabetes or other conditions, such as hypertension, that did not differentiate results by group were excluded. Studies including individuals with diabetes that did not specify type 1 or type 2 diabetes were included, as 90–95% of cases are type 2 diabetes (2). Studies looking at disincentives were not included. It is also important to note that studies examining the use of monetary approaches relevant to diabetes risk factors and prediabetes have been conducted; however, due to the primary focus on type 2 diabetes, these studies were not included in this review but have been covered in the final discussion, as they are considered key evidence for the field.
Data extraction included the first author’s last name, year of publication, study design, study aim, sample size, study population, intervention, monetary intervention approach, intervention duration, outcomes, and the country where the study was conducted. Critical appraisal and data quality followed standards outlined by the JBI critical appraisal checklist, which are appraised by study design (33–36). All authors were involved in the final article selections, which were done using a series of checklists to ensure all criteria for inclusion and quality were met for each article included in this review.
Results
Search and Study Selection
Figure 1 shows the PRISMA diagram with the records identified by database, including a manual search. The total number of records identified across each database and manual search was 988. A total of 163 duplicates were removed, and 825 records remained for screening. The title and abstract review resulted in exclusion of 795 records, leaving 30 to be assessed for eligibility using full-text review. Of the 30 records assessed, 17 were excluded for various reasons, leaving 13 articles for final synthesis.
Study Characteristics and Outcomes of Studies
Table 2 summarizes the study characteristics. Across the 13 studies, 5 were RCT, 5 were quasi-experimental, 2 were pre-post study designs, and 1 was a retrospective cohort. Ten of the studies were conducted in the U.S., with 1 study conducted in Jordan, 1 in Peru, and 1 in Singapore. Looking at the monetary approach across each study, eight used financial incentives, three used a form of income supplementation, one used cash transfers, and one used a combination of income supplementation with financial incentives. No studies were identified that used UBI in a type 2 diabetes population.
Study characteristics and intervention detail
| Author | Year | Study design | Aim | Sample size (n) | Study population demographics | Monetary approach | Intervention | Intervention duration | Monetary modality | Country |
|---|---|---|---|---|---|---|---|---|---|---|
| Bilger | 2021 | RCT | To examine the impact of financial incentives on diabetes outcomes among adults with type 2 diabetes and to examine if diabetes processes vs. clinical outcomes are differentially influenced by financial incentives | 240 | Adults aged 21–70 years with type 2 diabetes | Financial incentives | All patients received a glucose monitor, electronic medication bottle to measure medication-taking behavior, and a Fitbit zip to monitor physical activity. All participants also received text message reminders to engage in recommended self-management behaviors. There were two incentives groups, a process group and outcomes group. The process group received structured incentives based on completing self-care activities including blood glucose testing (Singapore [S] $3.50 weekly for testing 3× per week), medication (S$0.50 daily for taking medication), walking (S$1 daily for walking 8,000 per day). The outcomes group received incentives for achieving target glucose levels: 4–7 mmol/L (72–126 mg/dL) premeal, S$2 for 1× per week; S$7 for 2× per week; or S$14 for 3× per week. Both incentive groups could earn a maximum of S$14 ($10.36) per week. Note that at the time of this study, S$1.00 = US$0.74. The control group received a lump sum of S$75 at the end of 6 months for their participation. | 6 months | Voucher | Singapore |
| Bryce | 2017 | Pre-post | To examine the impact of a fruit and vegetable prescription program on diabetes clinical outcomes among patients with type 2 diabetes | 65 | Low-income adults aged ≥18 years with type 2 diabetes | Income supplementation + financial incentives | Patients receive $10 per week for up to 4 weeks, totaling $40 for use over the 13-week study period. Funds were given on a study debit card and could only be used for purchasing fresh fruits and vegetables. An additional $5 incentive was available if a goals sheet was completed at the initial visit. Funds were loaded at the farmers market site, and during this time, cooking demonstrations were held to reinforce healthy eating. CHW followed up with patients to discuss goals. | 13 weeks | Debit card | U.S. |
| Egede | 2021 | RCT | To incentivize behavior change | 60 | African American adults with type 2 diabetes; mean age 57.9 years; 72% women | Financial incentives | Diabetes education and skills training with home telemonitoring system + structured incentives. Group 1, single incentive at 3 months for absolute percentage drop in HbA1c from baseline to 3-month follow-up, up to $300. Group 2, single two-part incentive for uploading glucose measurements through home testing of glucose and absolute percentage drop in HbA1c from baseline to 3-month follow-up, up to $300. Group 3, single multiple-component incentive for uploading glucose measurements using home testing, attending weekly phone educational sessions, and absolute percentage drop in HbA1c from baseline to 3-month follow-up, up to $300. | 3 months | Money order | U.S. |
| Fenelon | 2022 | Quasi-experimental | To examine the impact of rental assistance on diabetes management among U.S. adults | 10,050 | Adults aged ≥45 years receiving and eligible to receive HUD rental assistance with a measure of HbA1c in NHANES | Income supplementation | Using HUD administrative data, diabetes management based on HbA1c cut points of 5.7% to ≤6.5%, >6.5%, and ≥9% among adults aged ≥45 years receiving housing assistance, defined as project-based housing or a housing voucher. Groups were compared with a waitlist comparison group, i.e., those who would enter HUD housing assistance within 2 years. | NA | Voucher; project housing | U.S. |
| Hager | 2023 | Quasi-experimental | To provide income assistance to facilitate healthy eating | 786 | Adults with diabetes, low income, risk of food insecurity | Income supplementation | Monthly $60 voucher to local grocery store. | 6 months | Voucher | U.S. |
| Long | 2012 | RCT | To examine the impact of peer mentoring vs. financial incentives on diabetes clinical outcomes | 118 | African American veterans with elevated HbA1c | Financial incentives | Patients were randomly assigned to 1 of 3 groups: peer mentoring, financial incentives, or usual care. Peer mentor groups were encouraged to talk weekly to set and review goals and provide support for diabetes management. Peers received $20 per month if meetings held weekly with participants. Participants in the financial incentive arm had the ability to receive $100 for dropping their baseline HbA1c by 1 point and could receive $200 for dropping their baseline HbA1c by 2 points or achieved an absolute value of ≤6.5%. | 6 months | Voucher redeemable for cash | U.S. |
| Lyles | 2021 | Quasi-experimental | To examine the impact of CCT plus diabetes health education (CHV + CCT) compared with diabetes health education alone (CHV) and unconditional CCT | 482 | Syrian refugee adults with type 2 diabetes living in poverty in Jordan | Cash transfer | CHV education session held quarterly for lifestyle and medication education, and service utilization for diabetes management + Quarterly cash transfer of $211 conditional upon purchasing medication and completion of health visits. UCT included monthly cash transfer based on family size ranging from $113 to $219. | 2 years | Cash transfer | Jordan |
| Miranda | 2019 | RCT | To examine the impact of individual financial incentives alone compared with having a supportive partner compared with shared incentives on type 2 diabetes outcomes | 44 | Adults aged 18–70 years with type 2 diabetes and overweight with BMI of 25–39.9 kg/m2 seen in local outpatient clinic | Financial incentives | Diabetes education paired with financial incentives for achieving goals over a 3-month period. For all groups the goals and structured incentives included 1) weight loss of 1 kg over a 2-week period ($25); 2) HbA1c decrease of <1% from baseline ($62); and 3) HbA1c decrease of ≥1% or reached levels of ≤6.5% compared with their baseline level ($124). Group 1 had individual incentives, and patients received financial incentive for meeting prespecified goals. Group 2 had mixed incentives-altruism, the patient identified a partner who provided support over the course of the study, and the patient alone received the financial incentives for meeting prespecified goals. Group 3 had mixed incentives-cooperation, the patient identified a partner who provided support over the course of the study, and if goals were achieved, the patient and partner were rewarded the incentives equally. | 3 months | Cash | Peru |
| Misra-Herbert | 2016 | Retrospective cohort | To examine the impact of a financial incentive employee wellness program in improving diabetes clinical outcomes | 1,092 | Adult employees of Cleveland Clinic participating in the EHP with diabetes, matched to nonemployees for comparison | Financial incentives | Employees participating in the disease management program for diabetes were offered $100 financial incentive for participating. One year after the program initiated, the incentive increased from $100 to $300. At the 2-year mark, the incentive was modified to a 30% discount on health insurance premium equal to $600 to $1,200 and was received for participation and achieving a clinical end point for HbA1c (<7%), LDL (<100 mg/dL), and blood pressure (SBP <130). A weight target was included but was individualized. The insurance discount was received the year after participation and achieving the clinical goals. | 5 years | Fixed- incentives modality not noted; goal incentives = discount in insurance premium | U.S. |
| Raiff | 2016 | Quasi-experimental | To examine combined text message reminders with financial incentives to increase medication taking for type 2 diabetes | 3 | Three adults aged 42, 43, and 51 years with type 2 diabetes seen in a local endocrinology clinic | Financial incentives | Patients were assigned an electronic medication bottle designed to track and provide a signal each time the bottle is opened, indicating medication is being taken. Patients customized the text messages they would receive as reminders only if medication taking was missed during a scheduled period. Incentives were vouchers that began at $1.00 and increased by $0.20 for each consecutive day of medication taking. For every third day of taking medications as prescribed, patients earned a $1.50 bonus. Missing a dose reset the incentive to $1.00. Total earnings over the course of the intervention could range between $73.50 and $84.10. The intervention period lasted 21–23 days. | 21–23 days | Voucher | U.S. |
| Sen | 2014 | RCT | To examine the impact of lottery-based financial incentives on home telemonitoring and clinical outcomes | 75 | Adults with diabetes | Financial incentives | All patients received a glucose monitor, blood pressure monitor, and scale. Each device electronically transmitted results to a study website. Patients were randomized to 1 of 3 groups: high-incentive group could receive $100 or $10 ($2.80 daily); low-incentive group could receive $50 or $5 ($1.40 daily); the third group had no incentives. Eligibility for incentives was based on having used all three home monitoring devices with results being uploaded to the study website daily. Text messages and emails were used to notify participants of the incentives, and if participants were eligible to win but had not used the devices, they were notified of the missed opportunity. Participant groups were blinded to each group, with each incentive group not being aware of the other and the control group not being aware that there were incentive groups. Incentive opportunities ended at 3 months, and groups were followed for an additional 3 months with no incentives. | 24 weeks | Not specified | U.S. |
| Tucker | 2004 | Quasi-experimental | To examine the impact of a gym-based financial incentive health program on diabetes clinical outcomes | 30 | Adults with type 2 diabetes | Financial incentives | All participants joined a gym and received points for attending group exercise sessions one time per week, attending primary care visits, laboratory visits, completing dilated retinal exams, keeping a diet and medication diary, attending the gym, participating in an American Diabetes Association fundraising walk, attending educational sessions held bimonthly, fruit consumption, blood pressure monitoring, fat loss, lean tissue gained, and completing an essay describing their experience in the program. At the end of the program, the points were tallied by gym staff, and awards with a cash prize were given: $3,000 for 1st; $1,000 for 2nd; $500 for 3rd; $750 for “Most Inspirational”; $750 for “Best Essay”; $250 for “Best Patient”; $1,000 for “Best Recipe”; $500 for most fat pounds lost. A male and female competitor were selected for each of the 1st, 2nd, and 3rd prizes. | 5 months | Cash | U.S. |
| Veldheer | 2021 | Pre-post | To examine the impact of fruit and vegetable prescription vouchers on diabetes and cardiovascular outcomes among adults with type 2 diabetes | 97 | Low-income adults age 18 and older with type 2 diabetes | Income supplementation | Patients received vouchers (equivalent to $1/household member/day for 28 days; range, $28−$140/month) to a local farmers market for 7 months paired with monthly diabetes education classes. | 7 months | Voucher | U.S. |
| Author | Year | Study design | Aim | Sample size (n) | Study population demographics | Monetary approach | Intervention | Intervention duration | Monetary modality | Country |
|---|---|---|---|---|---|---|---|---|---|---|
| Bilger | 2021 | RCT | To examine the impact of financial incentives on diabetes outcomes among adults with type 2 diabetes and to examine if diabetes processes vs. clinical outcomes are differentially influenced by financial incentives | 240 | Adults aged 21–70 years with type 2 diabetes | Financial incentives | All patients received a glucose monitor, electronic medication bottle to measure medication-taking behavior, and a Fitbit zip to monitor physical activity. All participants also received text message reminders to engage in recommended self-management behaviors. There were two incentives groups, a process group and outcomes group. The process group received structured incentives based on completing self-care activities including blood glucose testing (Singapore [S] $3.50 weekly for testing 3× per week), medication (S$0.50 daily for taking medication), walking (S$1 daily for walking 8,000 per day). The outcomes group received incentives for achieving target glucose levels: 4–7 mmol/L (72–126 mg/dL) premeal, S$2 for 1× per week; S$7 for 2× per week; or S$14 for 3× per week. Both incentive groups could earn a maximum of S$14 ($10.36) per week. Note that at the time of this study, S$1.00 = US$0.74. The control group received a lump sum of S$75 at the end of 6 months for their participation. | 6 months | Voucher | Singapore |
| Bryce | 2017 | Pre-post | To examine the impact of a fruit and vegetable prescription program on diabetes clinical outcomes among patients with type 2 diabetes | 65 | Low-income adults aged ≥18 years with type 2 diabetes | Income supplementation + financial incentives | Patients receive $10 per week for up to 4 weeks, totaling $40 for use over the 13-week study period. Funds were given on a study debit card and could only be used for purchasing fresh fruits and vegetables. An additional $5 incentive was available if a goals sheet was completed at the initial visit. Funds were loaded at the farmers market site, and during this time, cooking demonstrations were held to reinforce healthy eating. CHW followed up with patients to discuss goals. | 13 weeks | Debit card | U.S. |
| Egede | 2021 | RCT | To incentivize behavior change | 60 | African American adults with type 2 diabetes; mean age 57.9 years; 72% women | Financial incentives | Diabetes education and skills training with home telemonitoring system + structured incentives. Group 1, single incentive at 3 months for absolute percentage drop in HbA1c from baseline to 3-month follow-up, up to $300. Group 2, single two-part incentive for uploading glucose measurements through home testing of glucose and absolute percentage drop in HbA1c from baseline to 3-month follow-up, up to $300. Group 3, single multiple-component incentive for uploading glucose measurements using home testing, attending weekly phone educational sessions, and absolute percentage drop in HbA1c from baseline to 3-month follow-up, up to $300. | 3 months | Money order | U.S. |
| Fenelon | 2022 | Quasi-experimental | To examine the impact of rental assistance on diabetes management among U.S. adults | 10,050 | Adults aged ≥45 years receiving and eligible to receive HUD rental assistance with a measure of HbA1c in NHANES | Income supplementation | Using HUD administrative data, diabetes management based on HbA1c cut points of 5.7% to ≤6.5%, >6.5%, and ≥9% among adults aged ≥45 years receiving housing assistance, defined as project-based housing or a housing voucher. Groups were compared with a waitlist comparison group, i.e., those who would enter HUD housing assistance within 2 years. | NA | Voucher; project housing | U.S. |
| Hager | 2023 | Quasi-experimental | To provide income assistance to facilitate healthy eating | 786 | Adults with diabetes, low income, risk of food insecurity | Income supplementation | Monthly $60 voucher to local grocery store. | 6 months | Voucher | U.S. |
| Long | 2012 | RCT | To examine the impact of peer mentoring vs. financial incentives on diabetes clinical outcomes | 118 | African American veterans with elevated HbA1c | Financial incentives | Patients were randomly assigned to 1 of 3 groups: peer mentoring, financial incentives, or usual care. Peer mentor groups were encouraged to talk weekly to set and review goals and provide support for diabetes management. Peers received $20 per month if meetings held weekly with participants. Participants in the financial incentive arm had the ability to receive $100 for dropping their baseline HbA1c by 1 point and could receive $200 for dropping their baseline HbA1c by 2 points or achieved an absolute value of ≤6.5%. | 6 months | Voucher redeemable for cash | U.S. |
| Lyles | 2021 | Quasi-experimental | To examine the impact of CCT plus diabetes health education (CHV + CCT) compared with diabetes health education alone (CHV) and unconditional CCT | 482 | Syrian refugee adults with type 2 diabetes living in poverty in Jordan | Cash transfer | CHV education session held quarterly for lifestyle and medication education, and service utilization for diabetes management + Quarterly cash transfer of $211 conditional upon purchasing medication and completion of health visits. UCT included monthly cash transfer based on family size ranging from $113 to $219. | 2 years | Cash transfer | Jordan |
| Miranda | 2019 | RCT | To examine the impact of individual financial incentives alone compared with having a supportive partner compared with shared incentives on type 2 diabetes outcomes | 44 | Adults aged 18–70 years with type 2 diabetes and overweight with BMI of 25–39.9 kg/m2 seen in local outpatient clinic | Financial incentives | Diabetes education paired with financial incentives for achieving goals over a 3-month period. For all groups the goals and structured incentives included 1) weight loss of 1 kg over a 2-week period ($25); 2) HbA1c decrease of <1% from baseline ($62); and 3) HbA1c decrease of ≥1% or reached levels of ≤6.5% compared with their baseline level ($124). Group 1 had individual incentives, and patients received financial incentive for meeting prespecified goals. Group 2 had mixed incentives-altruism, the patient identified a partner who provided support over the course of the study, and the patient alone received the financial incentives for meeting prespecified goals. Group 3 had mixed incentives-cooperation, the patient identified a partner who provided support over the course of the study, and if goals were achieved, the patient and partner were rewarded the incentives equally. | 3 months | Cash | Peru |
| Misra-Herbert | 2016 | Retrospective cohort | To examine the impact of a financial incentive employee wellness program in improving diabetes clinical outcomes | 1,092 | Adult employees of Cleveland Clinic participating in the EHP with diabetes, matched to nonemployees for comparison | Financial incentives | Employees participating in the disease management program for diabetes were offered $100 financial incentive for participating. One year after the program initiated, the incentive increased from $100 to $300. At the 2-year mark, the incentive was modified to a 30% discount on health insurance premium equal to $600 to $1,200 and was received for participation and achieving a clinical end point for HbA1c (<7%), LDL (<100 mg/dL), and blood pressure (SBP <130). A weight target was included but was individualized. The insurance discount was received the year after participation and achieving the clinical goals. | 5 years | Fixed- incentives modality not noted; goal incentives = discount in insurance premium | U.S. |
| Raiff | 2016 | Quasi-experimental | To examine combined text message reminders with financial incentives to increase medication taking for type 2 diabetes | 3 | Three adults aged 42, 43, and 51 years with type 2 diabetes seen in a local endocrinology clinic | Financial incentives | Patients were assigned an electronic medication bottle designed to track and provide a signal each time the bottle is opened, indicating medication is being taken. Patients customized the text messages they would receive as reminders only if medication taking was missed during a scheduled period. Incentives were vouchers that began at $1.00 and increased by $0.20 for each consecutive day of medication taking. For every third day of taking medications as prescribed, patients earned a $1.50 bonus. Missing a dose reset the incentive to $1.00. Total earnings over the course of the intervention could range between $73.50 and $84.10. The intervention period lasted 21–23 days. | 21–23 days | Voucher | U.S. |
| Sen | 2014 | RCT | To examine the impact of lottery-based financial incentives on home telemonitoring and clinical outcomes | 75 | Adults with diabetes | Financial incentives | All patients received a glucose monitor, blood pressure monitor, and scale. Each device electronically transmitted results to a study website. Patients were randomized to 1 of 3 groups: high-incentive group could receive $100 or $10 ($2.80 daily); low-incentive group could receive $50 or $5 ($1.40 daily); the third group had no incentives. Eligibility for incentives was based on having used all three home monitoring devices with results being uploaded to the study website daily. Text messages and emails were used to notify participants of the incentives, and if participants were eligible to win but had not used the devices, they were notified of the missed opportunity. Participant groups were blinded to each group, with each incentive group not being aware of the other and the control group not being aware that there were incentive groups. Incentive opportunities ended at 3 months, and groups were followed for an additional 3 months with no incentives. | 24 weeks | Not specified | U.S. |
| Tucker | 2004 | Quasi-experimental | To examine the impact of a gym-based financial incentive health program on diabetes clinical outcomes | 30 | Adults with type 2 diabetes | Financial incentives | All participants joined a gym and received points for attending group exercise sessions one time per week, attending primary care visits, laboratory visits, completing dilated retinal exams, keeping a diet and medication diary, attending the gym, participating in an American Diabetes Association fundraising walk, attending educational sessions held bimonthly, fruit consumption, blood pressure monitoring, fat loss, lean tissue gained, and completing an essay describing their experience in the program. At the end of the program, the points were tallied by gym staff, and awards with a cash prize were given: $3,000 for 1st; $1,000 for 2nd; $500 for 3rd; $750 for “Most Inspirational”; $750 for “Best Essay”; $250 for “Best Patient”; $1,000 for “Best Recipe”; $500 for most fat pounds lost. A male and female competitor were selected for each of the 1st, 2nd, and 3rd prizes. | 5 months | Cash | U.S. |
| Veldheer | 2021 | Pre-post | To examine the impact of fruit and vegetable prescription vouchers on diabetes and cardiovascular outcomes among adults with type 2 diabetes | 97 | Low-income adults age 18 and older with type 2 diabetes | Income supplementation | Patients received vouchers (equivalent to $1/household member/day for 28 days; range, $28−$140/month) to a local farmers market for 7 months paired with monthly diabetes education classes. | 7 months | Voucher | U.S. |
Except where noted otherwise money values are in U.S. dollars. CHV, community health volunteers; CHW, community health workers; EHP, employee health plan; HUD, housing and urban development; NA, not applicable; NHANES, National Health and Nutrition Examination Survey; SBP, systolic blood pressure.
Table 3 shows impact across each outcome measured by each study and by monetary approach used. Of the 12 studies measuring HbA1c as an outcome, 9 demonstrated a statistically significant reduction in HbA1c across monetary approach and modality (37–45), with one showing clinically meaningful changes in HbA1c (46). Of the six studies that included blood pressure as an outcome, one demonstrated a statistically significant reduction in diastolic blood pressure (40). Of the six studies measuring BMI or weight as an outcome, three demonstrated statistically significant changes (40,41,44). Of the five studies measuring self-care behaviors, four demonstrated statistically significant improvements in self-care (40,41,43,47). Only two studies measured QOL as an outcome, and both demonstrated statistically significant improvements in QOL (44,47). Neither of the two studies measuring LDL as an outcome demonstrated statistically significant changes (42,44). Demographic information for each study is included in Table 2. Intervention duration ranged from 21 days to 5 years across studies.
Overall study findings by monetary approach and outcome
| First author | Year | Monetary approach | Outcome(s) measured | Statistical significance between monetary approach and diabetes outcomes | Monetary modality | Findings | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| HbA1c | Blood pressure | LDL | Self-care | BMI | QOL | ||||||
| Bilger | 2021 | Financial incentives | HbA1c; self-care behaviors; QOL | X | X | Voucher | There were no statistically significant differences HbA1c postintervention between the incentive groups and the usual care group (−0.31 [CI −0.67 to 0.06]). Self-care activities were significantly higher in the incentive group relative to control group for glucose testing (0.40 [CI 0.04–0.76]), medication taking (0.72 [CI 0.05–1.38]), and engaging in physical activity (1.12 [CI 0.38–1.86]). Similarly, glucose levels were significantly better for the incentive group relative to the control group (0.32 [CI 0.07–0.57]) as well as QOL (0.04 [CI 0.0–0.07]). | ||||
| Bryce | 2017 | Income supplementation + financial incentives | HbA1c; blood pressure | X | Debit card | Significant reduction in HbA1c was observed postintervention for study participants. Baseline HbA1c 9.5% (mean drop −0.71, P = 0.001). No significant change in blood pressure observed. | |||||
| Egede | 2021 | Financial incentives | HbA1c | X | Money order | Significant reduction in HbA1c was observed across each group at 3 months. Group 1 baseline HbA1c 10.0 (mean drop −1.25%; P = 0.002). Group 2 baseline HbA1c 9.9 (mean drop −1.73%; P < 0.001). Group 3 baseline HbA1c 10.4 (mean drop −1.74%; P < 0.001). | |||||
| Fenelon | 2022 | Income supplementation | HbA1c | X | Voucher; project housing | Results showed that for adults receiving housing assistance through project housing had a reduced likelihood of having HbA1c ≥9% by −3.7% compared with the waitlist group. Those receiving housing vouchers did not have statistically different reduced likelihood compared with the waitlist group. | |||||
| Hager | 2023 | Income supplementation | HbA1c; blood pressure; BMI | Voucher | No significant change in HbA1c, blood pressure, or BMI between treatment and control groups at 6 months. | ||||||
| Long | 2012 | Financial incentives | HbA1c | Voucher redeemable for cash | Significant reduction in HbA1c was observed for the peer mentor group (mean drop −1.07% [CI −1.84 to −0.31]), relative to control group. The financial incentive group was not statistically significant (mean drop of −0.45% [CI −1.23 to 0.32]) relative to control group. | ||||||
| Lyles | 2021 | Cash transfer | HbA1c; blood pressure; BMI; self-care behaviors | X | X | X | X | Cash transfer | Significant drop in HbA1c was observed for all 3 groups, i.e., CHV + CCT group (−0.5%; P < 0.001), CHV only group (−0.7%; P < 0.001), and UCT group (−0.2%; P = 0.028). Significant drop in diastolic blood pressure was observed in the CHV + CCT group (−3.3 mmHg; P = 0.001) and the CHV-only group (−2.5 mmHg; P = 0.048). Significant decrease in BMI was observed for CHV + CCT group only (−1.0 kg/m2; P = 0.005). Significant increases in medication taking were observed in the CHV + CCT group only (6.8%; P = 0.004). | ||
| Miranda | 2019 | Financial incentives | HbA1c; BMI; self-care behaviors | X | X | X | Cash | Significant reduction in HbA1c was observed across each group at 3 months. Group 1 baseline HbA1c 8.5 (mean drop −1.4%; P < 0.05). Group 2 baseline HbA1c 7.9 (mean drop −0.9%; P < 0.05). Group 3 baseline HbA1c 8.2 (mean drop −1.1%; P < 0.05). Significant reduction in BMI was observed for group 1 only at 3 months (−2.9; P < 0.05). Significant improvements in self-management behaviors were observed for group 1 and group 2 at 3 months relative to baseline (P < 0.05). Diet significantly improved across all 3 groups at 3 months (P < 0.05). Physical activity significantly improved for group 2 only at 3 months (P < 0.05). | |||
| Misra-Herbert | 2016 | Financial incentives | HbA1c; blood pressure; LDL | X | Fixed incentives modality not specified; goal incentives = discount in insurance premium | Significant reduction in HbA1c was observed in employees participating in the program compared with nonemployees, with a yearly mean change of −0.05 noted (P < 0.001). Significant changes in LDL and weight with a yearly mean change of −3.20 (P < 0.001) for the employee group and −2.35 (P < 0.001) for the nonemployee comparison group were noted as well as significant change in weight with a yearly mean change of −0.73 (P < 0.001) for the employee group and −0.37 (P < 0.001) for the nonemployee group. The fixed incentives compared with the goal-oriented incentives was not significantly associated with the change in HbA1c. | |||||
| Raiff | 2016 | Financial incentives | Self-care; medication | Voucher | Compared with baseline, medication taking showed a documented increase for 2 participants from 50% to 95% and 91%, respectively. For the third participant, baseline medication taking increased from 0 to 71%. Statistical analysis was not completed. | ||||||
| Sen | 2014 | Financial incentives | Self-care; HbA1c, blood pressure, BMI | X | X | Not specified | Rates of telemonitoring were significantly higher in both incentive groups relative to control group during the intervention period (81% low incentive vs. 58% control; P = 0.007) (77% high incentive vs. 58% control; P = 0.02). At month 4 after incentives ended, the high-incentive group telemonitoring dropped and was no longer different from that of control group. The level for the low-incentive group remained significantly higher relative to that of the control group through month 6. A significant change in HbA1c was observed from baseline to postintervention (−1.5%, CI −2.4 to −0.5, low incentive; −1.2%, CI −2.1 to −0.3, high incentive) but was not significantly different from that of the control group. No statistically significant changes were observed in blood pressure or BMI across groups. | ||||
| Tucker | 2004 | Financial incentives | HbA1c; weight loss; LDL; QOL | X | X | X | Cash | Significant reduction in HbA1c was observed for participants, mean change of 1.5 (P = 0.022). Significant reduction in weight was also observed, mean change of 16 lbs (P < 0.001). Significant improvements in QOL were also observed for physical component score, mean improvement of 4.3 points (P = 0.0012). | |||
| Veldheer | 2021 | Income supplementation | HbA1c; blood pressure; BMI | X | Voucher | Significant reductions in HbA1c were observed at follow-up compared with baseline (−1.3%; P < 0.001). Lower HbA1c at follow-up was associated with voucher redemption, as was high blood pressure. | |||||
| First author | Year | Monetary approach | Outcome(s) measured | Statistical significance between monetary approach and diabetes outcomes | Monetary modality | Findings | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| HbA1c | Blood pressure | LDL | Self-care | BMI | QOL | ||||||
| Bilger | 2021 | Financial incentives | HbA1c; self-care behaviors; QOL | X | X | Voucher | There were no statistically significant differences HbA1c postintervention between the incentive groups and the usual care group (−0.31 [CI −0.67 to 0.06]). Self-care activities were significantly higher in the incentive group relative to control group for glucose testing (0.40 [CI 0.04–0.76]), medication taking (0.72 [CI 0.05–1.38]), and engaging in physical activity (1.12 [CI 0.38–1.86]). Similarly, glucose levels were significantly better for the incentive group relative to the control group (0.32 [CI 0.07–0.57]) as well as QOL (0.04 [CI 0.0–0.07]). | ||||
| Bryce | 2017 | Income supplementation + financial incentives | HbA1c; blood pressure | X | Debit card | Significant reduction in HbA1c was observed postintervention for study participants. Baseline HbA1c 9.5% (mean drop −0.71, P = 0.001). No significant change in blood pressure observed. | |||||
| Egede | 2021 | Financial incentives | HbA1c | X | Money order | Significant reduction in HbA1c was observed across each group at 3 months. Group 1 baseline HbA1c 10.0 (mean drop −1.25%; P = 0.002). Group 2 baseline HbA1c 9.9 (mean drop −1.73%; P < 0.001). Group 3 baseline HbA1c 10.4 (mean drop −1.74%; P < 0.001). | |||||
| Fenelon | 2022 | Income supplementation | HbA1c | X | Voucher; project housing | Results showed that for adults receiving housing assistance through project housing had a reduced likelihood of having HbA1c ≥9% by −3.7% compared with the waitlist group. Those receiving housing vouchers did not have statistically different reduced likelihood compared with the waitlist group. | |||||
| Hager | 2023 | Income supplementation | HbA1c; blood pressure; BMI | Voucher | No significant change in HbA1c, blood pressure, or BMI between treatment and control groups at 6 months. | ||||||
| Long | 2012 | Financial incentives | HbA1c | Voucher redeemable for cash | Significant reduction in HbA1c was observed for the peer mentor group (mean drop −1.07% [CI −1.84 to −0.31]), relative to control group. The financial incentive group was not statistically significant (mean drop of −0.45% [CI −1.23 to 0.32]) relative to control group. | ||||||
| Lyles | 2021 | Cash transfer | HbA1c; blood pressure; BMI; self-care behaviors | X | X | X | X | Cash transfer | Significant drop in HbA1c was observed for all 3 groups, i.e., CHV + CCT group (−0.5%; P < 0.001), CHV only group (−0.7%; P < 0.001), and UCT group (−0.2%; P = 0.028). Significant drop in diastolic blood pressure was observed in the CHV + CCT group (−3.3 mmHg; P = 0.001) and the CHV-only group (−2.5 mmHg; P = 0.048). Significant decrease in BMI was observed for CHV + CCT group only (−1.0 kg/m2; P = 0.005). Significant increases in medication taking were observed in the CHV + CCT group only (6.8%; P = 0.004). | ||
| Miranda | 2019 | Financial incentives | HbA1c; BMI; self-care behaviors | X | X | X | Cash | Significant reduction in HbA1c was observed across each group at 3 months. Group 1 baseline HbA1c 8.5 (mean drop −1.4%; P < 0.05). Group 2 baseline HbA1c 7.9 (mean drop −0.9%; P < 0.05). Group 3 baseline HbA1c 8.2 (mean drop −1.1%; P < 0.05). Significant reduction in BMI was observed for group 1 only at 3 months (−2.9; P < 0.05). Significant improvements in self-management behaviors were observed for group 1 and group 2 at 3 months relative to baseline (P < 0.05). Diet significantly improved across all 3 groups at 3 months (P < 0.05). Physical activity significantly improved for group 2 only at 3 months (P < 0.05). | |||
| Misra-Herbert | 2016 | Financial incentives | HbA1c; blood pressure; LDL | X | Fixed incentives modality not specified; goal incentives = discount in insurance premium | Significant reduction in HbA1c was observed in employees participating in the program compared with nonemployees, with a yearly mean change of −0.05 noted (P < 0.001). Significant changes in LDL and weight with a yearly mean change of −3.20 (P < 0.001) for the employee group and −2.35 (P < 0.001) for the nonemployee comparison group were noted as well as significant change in weight with a yearly mean change of −0.73 (P < 0.001) for the employee group and −0.37 (P < 0.001) for the nonemployee group. The fixed incentives compared with the goal-oriented incentives was not significantly associated with the change in HbA1c. | |||||
| Raiff | 2016 | Financial incentives | Self-care; medication | Voucher | Compared with baseline, medication taking showed a documented increase for 2 participants from 50% to 95% and 91%, respectively. For the third participant, baseline medication taking increased from 0 to 71%. Statistical analysis was not completed. | ||||||
| Sen | 2014 | Financial incentives | Self-care; HbA1c, blood pressure, BMI | X | X | Not specified | Rates of telemonitoring were significantly higher in both incentive groups relative to control group during the intervention period (81% low incentive vs. 58% control; P = 0.007) (77% high incentive vs. 58% control; P = 0.02). At month 4 after incentives ended, the high-incentive group telemonitoring dropped and was no longer different from that of control group. The level for the low-incentive group remained significantly higher relative to that of the control group through month 6. A significant change in HbA1c was observed from baseline to postintervention (−1.5%, CI −2.4 to −0.5, low incentive; −1.2%, CI −2.1 to −0.3, high incentive) but was not significantly different from that of the control group. No statistically significant changes were observed in blood pressure or BMI across groups. | ||||
| Tucker | 2004 | Financial incentives | HbA1c; weight loss; LDL; QOL | X | X | X | Cash | Significant reduction in HbA1c was observed for participants, mean change of 1.5 (P = 0.022). Significant reduction in weight was also observed, mean change of 16 lbs (P < 0.001). Significant improvements in QOL were also observed for physical component score, mean improvement of 4.3 points (P = 0.0012). | |||
| Veldheer | 2021 | Income supplementation | HbA1c; blood pressure; BMI | X | Voucher | Significant reductions in HbA1c were observed at follow-up compared with baseline (−1.3%; P < 0.001). Lower HbA1c at follow-up was associated with voucher redemption, as was high blood pressure. | |||||
Summary of Evidence by Monetary Approach, Intervention Features, and Glycemic Control
Financial Incentives
Eight studies used financial incentives as the primary monetary intervention approach (38,41–44,46–48). Of these, five combined financial incentives with an additional feature, such as text message reminders (46–48), education (41–42), or home telemonitoring (43). Two layered multiple components onto structured financial incentives, including education, skills, and home telemonitoring (38), as well as tailored gym exercise sessions that included education and skills (44). Of interest, the studies layering multiple components onto financial incentives demonstrated the largest change in HbA1c among intervention participants, 1.5% (44) and 1.7% (38). The study that included diabetes education in addition to financial incentives saw a 1.4% change in HbA1c (41), and the study that included home telemonitoring with financial incentives saw a 1.2% change in HbA1c from baseline to follow-up (43), all statistically significant. The two studies that included text messaging did not demonstrate a statistical change in HbA1c (47,48). Another study found financial incentives tied to participation in an employee health program yielded a small, but significant, 0.19% relative decline in HbA1c in the first year of participation compared with nonemployee control individuals (42). While these comparisons are limited to qualitative synthesis, they suggest the importance of multicomponent interventions layered upon structured financial incentives based on HbA1c and underscore the role of diabetes education, skills training, and home telemonitoring as mechanisms to achieve optimal outcomes when combined with incentives across diverse populations.
Income Supplementation
Three studies used income supplementation as a primary monetary intervention approach (39,45,49). The first study found no significant differences in HbA1c levels between individuals with project-based housing or those with housing vouchers and control individuals (defined as people in a waitlist group) (39). However, receipt of project-based housing assistance was associated with a 3.7% lower likelihood of having uncontrolled diabetes (defined as HbA1c ≥9.0) compared with levels for the control group (39). A second study provided diabetes self-management education monthly along with the income supplementation for use at a local farmers market (45). Postintervention, participant mean HbA1c dropped by 1.3% from baseline (45). Another study provided monthly income supplementation in the form of grocery store vouchers and began the study with food label education, but this component was suspended early on due to the onset of the coronavirus 2019 (COVID-19) pandemic (49). Results did not demonstrate any statistically significant changes in HbA1c (49). The evidence presented here further highlights the importance of layering diabetes education onto a monetary approach as opposed to a single monetary approach with no educational and skills target tied to an outcome.
Cash Transfer Programs
This review found only one study designed to examine the use of cash transfers on diabetes outcomes (40). Using a multicomponent approach consistent with cash transfer designs, the conditionality in this study included spending funds on health service visits and prescription medication refills (40). In addition, participants received regular education and lifestyle education from a community health worker as well as home visits. This study also included a non–cash transfer group where participants received the community health worker intervention without the cash transfers. Study findings demonstrated a 0.5% change in HbA1c postintervention among the CCT group, 0.2% change in HbA1c among the UCT group, and a 0.7% change in the education only group; all changes were statistically significant (40). These findings show the impact that cash transfers, both CCT and UCT, have on HbA1c. Of interest, the conditionality for the conditional group was tied to the use of funds on health services and medication. Detail was not provided on the use of funds for basic needs and how this compares with more traditional approaches where the funds are conditional on attendance rather than how the funds are used. This study shows that cash transfers and intensive diabetes education and skills training accompanied by home visits are effective at reducing HbA1c and represents one of the first studies done using cash transfers directly tied to diabetes outcomes with a key focus on utilization. Further research is needed to understand if cash transfers conditional on education with no restrictions in use of funds are effective at improving HbA1c among diabetes populations.
Combined Monetary Approaches
One study used a combination of monetary approaches (37). Combining income supplementation with a small financial incentive tied to creating goals, this study demonstrated a 0.7% drop in HbA1c at follow-up among participants (37). Consistent with income supplementation, funds were provided up front and, in this study, could only be used to purchase fresh fruits and vegetables. In addition to the income supplementation and the financial incentive, study participants received cooking demonstrations and follow-up with community health workers to assess progress on goals (37). As a multicomponent intervention using combined monetary approaches with skills training and goal follow-up, this study demonstrates the effectiveness of using multiple strategies tied to behavior in diabetes.
Discussion and Implications
This is among the first reviews to summarize the evidence for interventions that use monetary approaches to improve diabetes outcomes for type 2 diabetes. Using a robust and reproducible search, 13 studies were identified that examined the association of one or more monetary approaches with change in diabetes clinical outcomes. Among these 13 studies, 10 demonstrated significant improvements in one or more diabetes outcomes, including HbA1c, blood pressure, self-care behaviors, BMI, and QOL. This review addresses three important gaps in the field. First, this review provides clear definitions and theoretical underpinnings of monetary approaches to improving diabetes outcomes. Second, this review provides a systematic summary of evidence by monetary approach. Third, this review summarizes key metabolic and psychosocial outcomes critical for optimizing health using monetary approaches.
Findings from this review show that interventions that use monetary approaches are associated with improving diabetes outcomes. Of note, 10 of the 13 studies demonstrated statistically significant and clinically meaningful changes in HbA1c. Changes in HbA1c ranged from 0.19% to 1.74% for intervention incorporating financial incentives, 0.7% to 1.3% for interventions incorporating income supplementation, and 0.2% to 0.7% for the study incorporating cash transfers. Given a change in HbA1c of 0.5% is clinically meaningful and associated with decreases in long-term complications, these are clinically important findings. However, use of these approaches for diabetes interventions is relatively new. The evidence reviewed shows that financial incentives have been studied more extensively with interventions spanning the past 20 years, although income supplementation and cash transfers included in this review are more recent, with the earliest publications being in 2017 and 2021, respectively. Of note, no studies were identified in this review that examined the impact of UBI on diabetes clinical outcomes.
Numerous studies have been conducted to examine the effect of cash transfers, namely, UCTs, on general self-reported health and social risk factors (50–52), with preliminary findings showing marginal to no impact (53–56). A white paper released during the writing of this review describes one of the largest randomized experiments conducted using UCTs over a 3-year period in the U.S. (57). Randomizing 3,000 adults, 1,000 to the UCT group and 2,000 to the control group, the OpenResearch Unconditional Income Study (ORUS) (58) examined self-reported health and social risk as well as key biological outcomes, including HbA1c, blood pressure, and cholesterol (57). At the end of the study period, compared with baseline, no statistically significant changes in biological outcomes were observed. However, improvements in self-reported health and key social risk factors such as food insecurity were reported after the first year of the program but were not sustained in the 3-year period (57).
Given the evidence base, targeted monetary interventions (i.e., focused on behaviors or conditionalities tied to disease outcomes) may be effective at improving outcomes for chronic disease conditions such as diabetes; however, future studies should consider comparison between different monetary approaches using designs that will allow identification of effective strategies. For example, comparative effectiveness and factorial designs will be helpful in identifying the separate and combined efficacy of strategies. In addition, incorporating cost analyses that assess prevention of complications and utilization can help understand cost-effectiveness of monetary approaches and inform policy recommendations. Finally, heterogeneity of effect should be considered to understand if effectiveness varies by population characteristics such as income.
Limitations
Several limitations warrant consideration. First, this review was restricted to publications in English; therefore, studies examining the impact of a monetary interventions published in other languages were excluded. Second, the scope of the search was limited to the literature indexed in Ovid MEDLINE, PubMed, Scopus, and CINAHL, as well as a manual search. Therefore, articles not represented in these databases, such as white papers and conference reports, may have been excluded. In addition, this review did not include studies that examined the role of disincentives on behavior change for type 2 diabetes. While this is consistent with financial incentive interventions in the field (12), additional work to explore the role of disincentives from a behavioral economic standpoint may be warranted. Finally, this scoping review is considered narrative, and a meta-analysis could not be conducted to estimate the magnitude of effect given the wide differences in intervention strategies.
Conclusions
Based on this review of the evidence, financial incentives, income supplementation, and cash transfers are important intervention strategies for improving diabetes outcomes in adults. Future work is needed to examine the longitudinal effect of individual and combined approaches on improving diabetes outcomes. In addition, examining the effect of UBI approaches on diabetes outcomes will be important as population-based approaches are developed to reduce the overall burden of disease. Finally, as financial incentives, income supplementation, and cash transfers are theoretically and structurally different, pathways identifying the underlying mechanisms of change are greatly needed to advance the field.
This article is featured in a podcast available at diabetesjournals.org/care/pages/diabetes_care_on_air.
Article Information
Funding. This study was partially supported by the National Institute of Diabetes and Digestive and Kidney Diseases (R01DK118038 and R01DK120861, principal investigator L.E.E.; K01DK131319, principal investigator J.A.C.), National Institute on Minority Health and Health Disparities (R01MD013826, principal investigators L.E.E. and R.J.W.; R01MD018012, principal investigators L.E.E. and S.L.; and R01MD017574, principal investigators L.E.E. and S.L.).
Duality of Interest. No potential conflicts of interest relevant to this article were reported.
Author Contributions. L.E.E. conceptualized the study. J.A.C., R.J.W., and S.L. drafted the manuscript. All authors were involved in critical revision of manuscript content. The final manuscript was approved by all authors.
Handling Editors. The journal editors responsible for overseeing the review of the manuscript were Steven E. Kahn and Alka M. Kanaya.
