The effects of targeting improvements in urine glucose on metabolic control in children with insulin dependent diabetes.

Epstein et al. (1981) set up a home program for kids with insulin-dependent diabetes. Parents gave points and praise each time the child did a urine glucose test. The team tracked testing and metabolic control across several kids using a multiple-baseline design.

Testing rates shot up and stayed high for every child. Yet lab numbers that show true metabolic control did not budge. The kids followed the new rule, but their bodies did not improve.

Jones et al. (1977) saw the opposite pattern. A token economy for teens on hemodialysis cut weight gain and also fixed blood chemistry. The difference: dialysis rewards were tied straight to the medical outcome (weight), not just the testing act.

McLean et al. (1983) used a similar home token system for everyday child compliance and saw big, lasting gains. Their task was social, not medical, so a pure behavior target was enough.

Lancioni et al. (2000) later helped an adult with memory loss use a 54-step checklist for blood-glucose checks. Her accuracy improved, but the study did not report metabolic change, mirroring the 1981 mixed result.

You can lock in a health behavior with points and praise, but that alone may not change the clinical number you really care about. Tie the reward to the outcome variable—like weight for dialysis—when possible. If you can’t, pair the behavior program with medical follow-up so the team can adjust insulin, diet, or meds while the child keeps testing.