Thompson 2012.

Study characteristics
Methods	Study design: RCT, 2 arms Geographic location: Maryland, USA Ethical approval: yes Recruitment setting: urban hospital‐based academic paediatric clinic Method of recruitment: 2 trained bilingual, bicultural research assistants recruited parents in the clinic waiting room; interested parents were consented by the use of an oral consent process. Length of follow‐up: no follow‐up Dropouts: no dropouts A priori calculation of effect size/power?: yes
Participants	Description: low‐income Spanish‐speaking parents of infants and toddlers Health topic Child nutrition and feeding Inclusion criteria Spanish‐speaking self‐reported Latino adults who were the primary caregiver to a child < 3 years Exclusion criteria Parents who had a child < 3 years with significant medical issues requiring special nutritional or feeding needs Intervention group Nutrition education via touchscreen (80 randomised and analysed) Control group Usual care (80 randomised and analysed*) Note: 2 participants in the control group were excluded from the analysis because they were missing any responses to the knowledge questionnaire. However, these participants were included in the analysis for the secondary outcome 'planned changes in behaviour'. PROGRESS‐Plus Place of residence: urban, USA Time living in host country (years), mean (n = 158): 6.02 Race/ethnicity: Latinos/Latinas Gender (n = 148): Intervention: 94.0% female Control: 91.0% female Education (years) (n = 159): 41.0% 6 y or less, 51.0% 7 to 12 y, 8.0% some or all of university degree Socioeconomic status/income: "low‐income" population (Thompson 2012) Health insurance: "More than 95% of clinic patients are publicly insured" (Thompson 2012, p. 413) Social capital (number of children), mean: 2.3 Age (years), mean: 27.55 Health literacy (baseline) Not measured
Interventions	Intervention: nutrition education via touchscreen Theoretical framework: behavioural, cognitive and humanistic learning theories, Health Belief Model, cultural targeting strategies Description: the intervention group members viewed 5 culturally and linguistically adapted modules on nutrition and feeding presented on an interactive platform using a touchscreen computer. The modules contained a series of short educational messages and included text, pictures and audio material that accounted for the educational levels and health literacy of the participants. The modules were interactive, meaning questions requiring participants' responses with feedback given. Content was partly tailored based upon these responses. Intervention provider: not applicable Delivery method/mode: 1 individual web‐based session (interactive touchscreen computer, 5 modules of 2 to 8 min, total duration approximately 25 min) Language of delivery: language concordant Format: partly tailored (algorithm‐based) Setting/location: semi‐private office setting Consumer involvement: no Comparator Type: usual care (no additional intervention) Description: participants in the control group did not receive any intervention.
Outcomes	Outcomes assessed in the study: parental nutrition and feeding knowledge, planned changes in behaviour Outcomes considered in this review Health literacy Apply (behaviour intent) (planned changes in behaviour) Health‐related knowledge (parental nutrition and feeding knowledge) Methods of assessing outcomes Face‐to‐face orally administered questionnaires by trained bilingual research assistants Parental nutrition and feeding knowledge: 19 questions including 12 true/false questions and 7 multiple choice questions (4 options) related to breastfeeding (5 questions), formula (3 questions), solid foods (3 questions), milk (4 questions) and juice (4 questions), 0 to 19, higher score is better Planned changes in behaviour: 3 questions including 1 question related to planned changes in behaviour on the basis of the lessons learned ("yes"/"perhaps"/"no"), 1 open‐ended question on exactly what behaviours participants want to change, and 1 question on plans about talking to the child's doctor, family or friends about the information (yes, probably, no) Language of assessment: language concordant Translation procedure (if necessary): back‐translation technique Reliability/validity: developed for the study, no psychometric properties reported Timing of outcome assessment: baseline, short‐term (immediately after intervention)
Health literacy	Definition: not reported Health literacy components addressed by the intervention Prerequisites and tools Knowledge Motivation Steps of information processing Understand Appraise Apply Health domain: disease prevention (prevent childhood diseases through nutritional failure)
Notes	Trial ID: NCT01272492 Funding: funding was provided by Johns Hopkins University.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Participants were randomized by the use of a block randomization process. We used block randomization, 10‐per‐block, to prevent sample size imbalances which could affect the study’s power. At the start of the trial, an opaque container was filled with 10 envelopes with equal representation of intervention and control assignments. The research assistant removed an envelope from this container to determine each participant’s group assignment. After ten participants, she repeated the process."
Allocation concealment (selection bias)	Low risk	The randomisation procedure used indicates a low risk of selection bias.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Blinding of participants and personnel was not reported and behaviour intent was subjectively measured. It is unclear whether the results were affected.
Blinding of outcome assessment (detection bias) subjective outcome measures	Unclear risk	Participants were probably not blinded to group allocation and behaviour intent was assessed using a verbally administered questionnaire.
Blinding of outcome assessment (detection bias) objective outcome measures	Low risk	Unclear blinding but knowledge was objectively measured and not subject to interpretation.
Incomplete outcome data (attrition bias) All outcomes	Low risk	"Only 2 participants were missing any responses to the knowledge questions.These individuals were not included in the analyses for the total summed knowledge score and the breastfeeding domain‐specific summed knowledge score." No participant was lost to follow‐up and only 2 participants were excluded from the analysis due to missing responses.
Selective reporting (reporting bias)	Low risk	All prespecified outcomes reported at clinicaltrials.gov are reported in the published reports.