Skip to main content
. 2020 May 28;8(5):e15111. doi: 10.2196/15111

Table 8.

Outcomes and bias of trials with a focus on general health, preventive health, health beliefs, and other topics.

Reference Participant age (years), mean (SD) Attrition rate Main outcomes Bias tool Bias rating Bias reasoning
Moniz et al (2015) [49] 24.0 (4.5) 8% (13/171) Participants agreed that receiving text messages changed their beliefs about targeted preventive health behaviors:

  • Smoking (50%)

  • Sexually transmitted disease prevention (72%)

  • Prenatal vitamins (83%)

  • Seat belt use (68%)

  • Nutritious food intake (84%)

  • Breastfeeding (68%)

 NIH QATa Fair risk No before/after or multiple measurements taken
Dalrymple et al (2013) [50] Unclear 84% (26/31) for posttest; 35% (11/31) for any monthly form

  • 100% agreed “I tried to eat better for myself and the baby.”

  • 60% agreed “I understood what was happening to my body better.”

 NIH QAT High risk No before/after or multiple measurements taken, small sample size, high attrition
Bush (2017) [51] Unclear Unclear

  • Significant association between app use and completion of a prenatal visit at least 6 months before delivery (ORb 1.76; P=.02)

  • Borderline significant association between app use and low birth weight (OR 0.25; P=.06)

  • No association between app use and cesarean delivery or NICUc admission

 NIH QAT High risk Used a comparison that was not randomly selected (self-selected app users)
Krishnamurti (2017) [52] Median 24, range (18-35) 0% (0/16)

  • Intervention users reported higher intention to breastfeed at 2 months (t13=−4.16; P=.001) and 3 months (t15=−2.76; P=.01)

  • No statistical significance in intention to use prenatal vitamins

  • Clinic attendance rate was higher in participants than nonparticipant clinic patients (84% vs 50%)

  • Attendance was even higher (89%) among those who scheduled free Uber transportation

 NIH QAT High risk Sample size too low
Ledford (2016) [53] 29.29 (4.8) for IGd vs 29.37 (4.83) for CGe 27% (46/173)

  • Mobile group reported more frequent use (P=.04) and greater patient activation (P=.02) than the notebook group

  • No difference in biometrics including blood pressure control, weight gain, delivery outcomes

 Cochrane ROBTf Fair risk Unclear how randomization occurred, patients not blinded
Evans et al (2014) [54] 26.53 (SD not noted) 51.3% (484/943)

  • Significantly more of the intervention group agreed that “Taking prenatal vitamins will improve the health of my developing baby” (OR 1.91; P=.02)

  • No difference in outcomes including self-reported smoking, consumption of alcoholic beverages or fruit and vegetable consumption

 Cochrane ROBT Fair risk Selective reporting, high attrition
Evans et al (2012) [55] 27.6 (SD not noted) 27% (33/123)

  • Significantly more of the intervention group agreed that “I am prepared to be a new mother” (OR 2.73; P=.04)

  • No difference in outcomes including beliefs that smoking will harm the developing baby, that drinking alcohol will harm the developing baby, and that taking prenatal vitamins will improve the health of the developing baby

 Cochrane ROBT Fair risk Unclear blinding of participants and personnel; incomplete outcome data
Takeuchi (2016) [56] 32.7 (4.59) for IG vs 32.5 (4.18) for CG 40% (65/161)

  • No difference in practice of perineal massage, perineal lacerations, or episiotomy rates

 Cochrane ROBT High risk High attrition rate, self-assessment by unblinded participants, unclear randomization

aNIH QAT: NIH Quality Assessment Tool.

bOR: odds ratio.

cNICU: neonatal intensive care unit.

dIG: intervention group.

eCG: control group.

fROBT: risk of bias tool.