Table 4.

Overall strength of the evidence.

Outcome	Studies (observations), n	Summary of findings	Consistency, directness, and precision	Limitations (including reporting bias)	Overall strength of evidence	Applicability
Weight	5 [1,2,29-31] (689)	Increased weight loss noted in all intervention groups but significant in only 3 studies. One high-quality study [30] (n=371) reported a significant difference in weight change between participants in the intervention and those in the control group	Consistent, indirect, and precise	Two studies had a high risk of bias for nonblinding of study personnel, participants, and outcome assessors. One study had a high risk of bias for incomplete data	Moderate	African American and Hispanic adults with obesity or morbid obesity, who are young and middle-aged adults, and have access to a mobile phone
Hemoglobin A1c	3 [1,12,28] (380)	Improvement in hemoglobin A1c in the intervention groups but only significant in one study	Consistent, indirect, and imprecise	Studies had a high risk of bias from the nonblinding of study personnel and participants. One study also had a high risk of bias from the nonblinding of outcome assessors and the other study had an unclear risk of bias for the same	Low	African American or Hispanic adults with poorly controlled diabetes and access to a mobile phone