Investigating Rewards and Deposit Contract Financial Incentives for Physical Activity Behavior Change Using a Smartphone App: Randomized Controlled Trial
Journal
Journal of Medical Internet Research
ISSN
1438-8871
Type
journal article
Date Issued
2022-10-06
Author(s)
de Buisonjé, David R
;
Reijnders, Thomas
;
Cohen Rodrigues, Talia R
;
Prabhakaran, Santhanam
;
;
Lipman, Stefan A
;
Bijmolt, Tammo H A
;
Breeman, Linda D
;
Janssen, Veronica R
;
Kraaijenhagen, Roderik A
;
Kemps, Hareld M C
;
Evers, Andrea W M
Abstract
Background:
Financial incentive interventions for improving physical activity have proven to be effective but costly. Deposit contracts (in which participants pledge their own money) could be an affordable alternative. In addition, deposit contracts may have superior effects by exploiting the power of loss aversion. Previous research has often operationalized deposit contracts through loss framing a financial reward (without requiring a deposit) to mimic the feelings of loss involved in a deposit contract.
Objective:
This study aimed to disentangle the effects of incurring actual losses (through self-funding a deposit contract) and loss framing. We investigated whether incentive conditions are more effective than a no-incentive control condition, whether deposit contracts have a lower uptake than financial rewards, whether deposit contracts are more effective than financial rewards, and whether loss frames are more effective than gain frames.
Methods:
Healthy participants (N=126) with an average age of 22.7 (SD 2.84) years participated in a 20-day physical activity intervention. They downloaded a smartphone app that provided them with a personalized physical activity goal and either required a €10 (at the time of writing: €1=US $0.98) deposit up front (which could be lost) or provided €10 as a reward, contingent on performance. Daily feedback on incentive earnings was provided and framed as either a loss or gain. We used a 2 (incentive type: deposit or reward) × 2 (feedback frame: gain or loss) between-subjects factorial design with a no-incentive control condition. Our primary outcome was the number of days participants achieved their goals. The uptake of the intervention was a secondary outcome.
Results:
Overall, financial incentive conditions (mean 13.10, SD 6.33 days goal achieved) had higher effectiveness than the control condition (mean 8.00, SD 5.65 days goal achieved; P=.002; ηp2=0.147). Deposit contracts had lower uptake (29/47, 62%) than rewards (50/50, 100%; P<.001; Cramer V=0.492). Furthermore, 2-way analysis of covariance showed that deposit contracts (mean 14.88, SD 6.40 days goal achieved) were not significantly more effective than rewards (mean 12.13, SD 6.17 days goal achieved; P=.17). Unexpectedly, loss frames (mean 10.50, SD 6.22 days goal achieved) were significantly less effective than gain frames (mean 14.67, SD 5.95 days goal achieved; P=.007; ηp2=0.155).
Conclusions:
Financial incentives help increase physical activity, but deposit contracts were not more effective than rewards. Although self-funded deposit contracts can be offered at low cost, low uptake is an important obstacle to large-scale implementation. Unexpectedly, loss framing was less effective than gain framing. Therefore, we urge further research on their boundary conditions before using loss-framed incentives in practice. Because of limited statistical power regarding some research questions, the results of this study should be interpreted with caution, and future work should be done to confirm these findings.
Financial incentive interventions for improving physical activity have proven to be effective but costly. Deposit contracts (in which participants pledge their own money) could be an affordable alternative. In addition, deposit contracts may have superior effects by exploiting the power of loss aversion. Previous research has often operationalized deposit contracts through loss framing a financial reward (without requiring a deposit) to mimic the feelings of loss involved in a deposit contract.
Objective:
This study aimed to disentangle the effects of incurring actual losses (through self-funding a deposit contract) and loss framing. We investigated whether incentive conditions are more effective than a no-incentive control condition, whether deposit contracts have a lower uptake than financial rewards, whether deposit contracts are more effective than financial rewards, and whether loss frames are more effective than gain frames.
Methods:
Healthy participants (N=126) with an average age of 22.7 (SD 2.84) years participated in a 20-day physical activity intervention. They downloaded a smartphone app that provided them with a personalized physical activity goal and either required a €10 (at the time of writing: €1=US $0.98) deposit up front (which could be lost) or provided €10 as a reward, contingent on performance. Daily feedback on incentive earnings was provided and framed as either a loss or gain. We used a 2 (incentive type: deposit or reward) × 2 (feedback frame: gain or loss) between-subjects factorial design with a no-incentive control condition. Our primary outcome was the number of days participants achieved their goals. The uptake of the intervention was a secondary outcome.
Results:
Overall, financial incentive conditions (mean 13.10, SD 6.33 days goal achieved) had higher effectiveness than the control condition (mean 8.00, SD 5.65 days goal achieved; P=.002; ηp2=0.147). Deposit contracts had lower uptake (29/47, 62%) than rewards (50/50, 100%; P<.001; Cramer V=0.492). Furthermore, 2-way analysis of covariance showed that deposit contracts (mean 14.88, SD 6.40 days goal achieved) were not significantly more effective than rewards (mean 12.13, SD 6.17 days goal achieved; P=.17). Unexpectedly, loss frames (mean 10.50, SD 6.22 days goal achieved) were significantly less effective than gain frames (mean 14.67, SD 5.95 days goal achieved; P=.007; ηp2=0.155).
Conclusions:
Financial incentives help increase physical activity, but deposit contracts were not more effective than rewards. Although self-funded deposit contracts can be offered at low cost, low uptake is an important obstacle to large-scale implementation. Unexpectedly, loss framing was less effective than gain framing. Therefore, we urge further research on their boundary conditions before using loss-framed incentives in practice. Because of limited statistical power regarding some research questions, the results of this study should be interpreted with caution, and future work should be done to confirm these findings.
Language
English
HSG Classification
contribution to scientific community
HSG Profile Area
SoM - Business Innovation
Refereed
Yes
Publisher
JMIR Publications
Volume
24
Number
10
Official URL
Division(s)
Eprints ID
267540
File(s)![Thumbnail Image]()
open.access
Name
de Buisonje et al 2022 Rewards and Financial Incentives for Digital PA-RCT.pdf
Size
426.53 KB
Format
Adobe PDF
Checksum (MD5)
a69f1b965a6167ac1392d5e073866a12