NC: Behavior Change Strategies (2007-2008)

Citation:
 
Study Design:
Class:
- Click here for explanation of classification scheme.
Quality Rating:
Research Purpose:

To systematically review evidence for the effects of financial contingencies in the behavioral treatment of obesity, focusing on the use of only guaranteed incentive schemes.

Inclusion Criteria:
  • Randomized controlled trial for treatment of obesity involving use of financial incentives as rewards contingent on weight loss or other behavior change
  • Outcome data includes weight change
  • Minimum follow-up of one year
  • Age ≥18 years
  • BMI ≥28kg/m2.
Exclusion Criteria:
  • None specified.
Description of Study Protocol:

Search Procedures

  • Searches performed in Medline, Embase, Cinahl, PsychINFO, Cochrane Database of Systematic Reviews, Cochrane Register of Controlled Clinical Trials and Sportdiscus
  • Regular citation alerts used (except in Cochrane where not possible)
  • Relevant journals searched: International Journal of Obesity, Addictive Behavior, Eating Behavior, Journal of Consulting and Clinical Psychology, Journal of Applied Behavior Analysis, Journal of Health Psychology and Professional Psychology
  • Search terms: Financial, contract(ing), Mone(tar)y, Pay(ment), reward, contingency, motivation, reinforcement, obesity, overweight
  • Searches for authors who had published in this field using preceding searches and reference lists
  • Full-text article of identified relevant articles retrieved
  • Full-text articles also retrieved when decision to include or exclude was not clear for abstract
  • Authors contacted for studies when results were to be reported at a later date.

Selection of Studies

  • Two reviewers independently involved in identifying and retrieving articles 
  • Both reviewers had to agree to relevance of the study to the review according to the a priori criteria.

Quality Assessment of studies

Consensus had to be reached by both reviewers on the quality of the study using a quality assessment from Avenell et al, 2004.

Design

Systematic review.

Interventions

Randomized controlled trials of behavioral treatments for obesity of at least one year that included financial incentives as rewards contingent on weight loss or other behavior change

Statistical Analysis

  • Studies entered into Review Manager
  • Statistical meta-analysis: To determine the effect size of financial incentives by calculating the weighted mean difference (MD) for weight change and 95% confidence interval (CI)
  • Individuals within-study groups differing only in the provision of financial incentives were compared.

Prue-specified Sub-Group Analysis to Compare

  • Duration of use of incentives above and below median intervention period
  • Monetary value of incentive (as a percentage of personal disposable income (PDI)) above and below the median of the maximum value
  • Reward based on weight change
  • Reward based on other behavior change
  • Reward for group performance
  • Reward for individual performance
  • Reward by psychologist
  • Reward by non-psychologist
  • Ethnic groups at high risk for metabolic syndrome with those who were not
  • Heterogeneity across studies: I2<50%: Fixed approach used; I2>59% random effects approach used.
Data Collection Summary:

Data Abstraction

  • Data abstracted using form from Avenell et al, 2004, by first author
  • Checked by second author
  • Discussion to reach consensus in areas of differences
  • Authors contacted for further details as necessary
  • Graphical images of weight data were scanned and analyzed when only graphical data was provided
  • Monetary value of incentive (maximum possible reward): Percentage of personal disposable income (PDI) at year of publication using published data for the appropriate country
  • All weight data converted to kilograms where necessary
  • Missing standard deviations (SD) computed using linear regression plot of the SD of the mean change in weight on the absolute mean change for weight already derived in Avenell: SD of weight change =5.915+(0.283x absolute value of mean change in weight). 
Description of Actual Data Sample:

Results of Search

  • Number of articles identified =45,589
  • Number identified as RCTs of obesity treatments using financial incentives =38
  • Number meeting criteria for inclusion in analysis =9
  • Number included in meta-analysis =7 (one article did not report weight change; the other article used a crossover method in delivery of refund).

Description of Included Studies

  • All were behavioral obesity treatments
  • Dates conducted: 1970s (two studies); 1980s (four studies); 1990s (three studies)
  • Source of recruitment not clear (one study)
  • Existing pool of volunteers used (one study)
  • Existing pool of volunteers compared with those fresh from media recruitment (one study)
  • Recruiting through media advertisement (six studies).

Participants

  • Mean age: 35.7 to 52.8 years
  • Mean Body Mass Index (BMI): 29.3 to 31.8kg/m2
  • More females in all studies except one all-male study
  • Socioeconomic status not indicated in most studies, but where indicated, female participants were of lower socioeconomic status than male participants
  • Ethnicity given in two studies: White: 71% to 98%
  • Smokers excluded from one study; smoking status not described in other studies.
Summary of Results:

Interventions

Financial incentives:

  • Financial incentives freely supplied (two studies)
  • Financial incentives provided from participants' deposited money (seven studies)
  • Refunds were made for weight loss, compliance with behavior change, or attendance at sessions
  • Some studies compared refund for weight change with refund for compliance with behavior change (three studies)
  • One study compared reward by psychologist with reward by a non-psychologist
  • Duration of use of incentives = eight weeks to 18 months; the financial incentive was usually discontinued after about 16-24 weeks or became less regular in the longer studies 
  • Monetary value of financial incentives =0.5% PDI to 8.8% PDI; median =3.1% PDI (When the incentive was used for less than or more than one year, estimation was performed to enable comparisons using simple pro rata calculations, and making allowance for situations where the size of the incentive payment changed over time)
  • All included studies coordinated by psychologists
  • Intervention groups received behavioral, diet and exercise advice

 Other Incentives

  • Some included provision of food (two studies)
  • Some included exercise trainers (three studies)

Follow-up and Attrition

  • Longest reported follow-up periods was 30 months
  • Lowest overall study rate of attrition was 1.1% at 24 months
  • Highest overall attrition rate was 57.9% at 13 months, where greater attrition (P<0.10) was observed when the weight loss contract was in place than when an attendance contract was in place.

Weight change in studies comparing use of financial incentives vs. no financial incentive

Financial Incentives vs. No Financial Incentives(N)

Mean WMD* Weight Change (kg)
(95% CI)

Heterogeneity
(Chi Square P)

Overall Effect Z (P)

12 months (352)

-0.4

(-1.6 to 0.8)

P=0.49

I2=0%

Z=0.64

P=0.52)

18 months (124)

 -0.7

(-2.5 to 1.1)

P=0.72

I2 = 0%

Z=0.81

P=0.42

30 months (48)

1.1

(-1.3 to 3.4)

P=0.77

I2=0%

Z=0.9

P=0.37

 * WMD: Weighted Mean Difference

Effects of use of financial incentives compared by percentage PDI, and compared without financial incentives

 

Mean WMD Weight Change (kg)

(95% CI)

Heterogeneity (Chi Square P)

 

Overall Effect Z (P)
Use of financial incentives <1.2% PDI vs. no financial incentive at 12 months (N=112)

0.0

(-1.5 to1.6)

P=0.37

I2=6.5%

Z=0.02

P=0.96

Use of financial incentives of 1.2% PDI or more vs. use of financial incentives <1.2% PDI at 12 months (N=56)

-0.7

(-41. to 2.7)

P=0.63

I2=0%

Z=0.41

P=0.68

Use of financial incentives from 1.2% PDI vs. no financial incentives at 12 months (N=140)

-1.1

(-3.1 to 0.9)

P=0.51

I2=0%

Z=1.07

P=0.28

Use of financial incentives from 0.5% PDI vs. no financial incentive at 18 months (N=124)

-0.7

(-2.5 to 1.1)

P=0.72

I2=0%

Z=0.81

P=0.42

Assuming Use of Financial Incentives for One Year in All Studies

  • Comparison of groups receiving financial incentives equivalent to less than 3.1% PDI to groups not receiving an financial incentive: WMD change in weight =-0.4kg (-1.7 to 0.9); 
  • Comparison of groups receiving 3.1% PDK and above to groups not receiving any financial incentive: Mean WMD change in weight =-0.9 (-2.8 to 1.1).

Use of Financial Incentives Above and Below the Median Intervention Period (16 weeks)

  • At 12 months, treatment with financial incentives for less than 16 weeks (the median intervention period) was associated with a WMD weight change of -0.8kg (-2.3 to 0.7kg) compared to treatment without financial incentives
  • Use of financial incentives for more than 16 weeks was associated with a WMD weight change of 0.5kg (-1.7 to 2.5kg).

Comparisons of Different Methods Delivering Reward Using Financial Incentives

 

Mean WMD Weight Change (kg)
(95% CI)

Chi Square

Overall Effect Z (P)

Reward for group performance vs reward for individual performance at 12 months (N=44)

-2.1

(-5.4 to 1.2)

P=0.77

I2=0%

Z=1.25

P=0.21)

Reward for weight loss vs reward for behavior change at 12 months (N=38)

1.0

(-1.5 to 3.4)

P=0.03

I2=71.5%

Z=0.75

P=0.46

Reward by psychologist vs reward by non-psychologist at 12 months (N=12)

3.0

(-8.6 to 2.5)

P=0.45

I2=0%

Z=1/06

P=0.29

Other Findings

  • There was insufficient data to compare the freely supplied incentive groups and the groups receiving refunded deposits at the same time
  • Comparing groups with the financial incentives provided by participants as refundable deposits with no financial incentive control groups at 12 months: WMD weight change =-0.5kg (-1.8 to 0.9)
  • Comparing groups with freely supplied financial incentives with control or comparison groups not receiving financial incentives at 18 months: WMD weight change =-0.7kg (-2.5 to 1.1) 
  • Subgroup analysis of ethnic groups at high risk of metabolic syndrome and of participants from lower socioeconomic groups was not performed because of insufficient data.

 

Author Conclusion:

The use of financial incentives does not significantly effect weight loss or maintenance at 12 and 18 months, although weak trends favor use of incentives greater than 1.2% PDI under certain situations including rewarding for behavior change rather than weight, rewarding based on group performance rather than for individual performance and reward delivered by non-psychologists rather than psychologists.

Funding Source:
Government: Scottish Executive Health Dept.
Reviewer Comments:
  • Study interventions were of relatively short duration (median =16 weeks)
  • Participants were only overweight or moderately obese, thus not generalizable to more obese populations
  • Participants primarily from U.S. (one from Canada), thus limiting generalizability.

 

 

Quality Criteria Checklist: Review Articles
Relevance Questions
  1. Will the answer if true, have a direct bearing on the health of patients? Yes
  2. Is the outcome or topic something that patients/clients/population groups would care about? Yes
  3. Is the problem addressed in the review one that is relevant to dietetics practice? Yes
  4. Will the information, if true, require a change in practice? Yes
 
Validity Questions
  1. Was the question for the review clearly focused and appropriate? Yes
  2. Was the search strategy used to locate relevant studies comprehensive? Were the databases searched and the search termsused described? Yes
  3. Were explicit methods used to select studies to include in the review? Were inclusion/exclusion criteria specified andappropriate? Wereselectionmethods unbiased? Yes
  4. Was there an appraisal of the quality and validity of studies included in the review? Were appraisal methodsspecified,appropriate, andreproducible? Yes
  5. Were specific treatments/interventions/exposures described? Were treatments similar enough to be combined? Yes
  6. Was the outcome of interest clearly indicated? Were other potential harms and benefits considered? Yes
  7. Were processes for data abstraction, synthesis, and analysis described? Were they applied consistently acrossstudies and groups? Was thereappropriate use of qualitative and/or quantitative synthesis? Was variation in findings among studies analyzed? Were heterogeneity issued considered? If data from studies were aggregated for meta-analysis, was the procedure described? Yes
  8. Are the results clearly presented in narrative and/or quantitative terms? If summary statistics are used, are levels ofsignificance and/or confidence intervals included? Yes
  9. Are conclusions supported by results with biases and limitations taken into consideration? Are limitations ofthe review identified anddiscussed? Yes
  10. Was bias due to the review's funding or sponsorship unlikely? Yes