Summary of evidence table for digital health intervention studies to assist pregnant smokers achieve abstinence
| Population: | Pregnant smokers | |||||
| Setting: | Any setting* | |||||
| Intervention: | Digital health† | |||||
| Comparison: | Usual care or control‡ | |||||
| Outcome | Anticipated absolute effects (95% CI) | Relative effect (95% CI)§ | No. of participants (studies) | Certainty of evidence (Grading of Recommendations Assessment, Development and Evaluation) | Comments | |
| Effect with usual care or control group | Effect with intervention | |||||
| Digital health interventions | ||||||
| Digital health interventions (biochemically validated) on point prevalence abstinence¶ in late pregnancy | 89 per 1000 | 98 per 1000 | 1.37 (0.90 to 2.07) | 2845 (8) | ⊕○○○** Very low | No heterogeneity with SMS interventions only but moderate for the combined types of digital health interventions. Only one smart phone application trial could be included for this meta-analysis and this contained a contingency or monitory reward component. |
| Digital health interventions (biochemically validated) on point prevalence abstinence at postpartum (1–3 months, 4–6 months and 7–18 months and unspecified) | 76 per 1000 | 104 per 1000 | 1.46 (1.05 to 2.92) | 2238 (5) | ⊕⊕○○†† Low | |
| Digital health interventions (biochemically validated) on continuous abstinence‡‡ in late pregnancy | 34 per 1000 | 57 per 1000 | 1.98 (1.08 to 3.64) | 2049 (4) | ⊕⊕⊕○§§ Moderate | |
| Digital health interventions (biochemically validated) on continuous abstinence at postpartum (1–3 months, 4–6 months and 7–18 months and unspecified) | 10 per 1000 | 28 per 1000 | 2.68 (0.73 to 9.79) | 613 (1) | See comment | Only one study78 with one measurement could be included for this outcome. |
*All included studies were conducted in high income countries (USA and UK) in English speaking populations, where recruitment took place through antenatal clinics or online such as social media. Settings for recruitment would not be considered to have a significant influence on this type of interventions and their respective outcomes.
†Digital health interventions included any intervention that can potentially be communicated through the current technology (eg, hand held smart phone). These included SMS interventions, smart phone applications, computer or web-based programmes.
‡Usual care often included brief counselling including the brief use of the 5As (ie, ask, advice, assess, assist and arrange) approach. In other studies, controls matched the intervention (eg, SMS intervention matched with SMS information to quit smoking, etc).
§Relative effect was calculated as relative risk where risk is treated as effect rather than risk and CI is CI for this calculated relative effect.
¶Point prevalence abstinence was either defined as point prevalence abstinence of various time periods or was abstinence at a point in time that was biochemically validated but was not specifically defined. This type of abstinence did not include continuous abstinence which is understood as abstinence for a period of time rather than a point in time.
**Very low certainty in evidence due to high risk of bias in at least three studies, inconsistency due to varying effect estimates and moderate heterogeneity, imprecision due to CI crossing value of 1 and possible publication bias.
††Low certainty in evidence due to high risk of bias with one study and imprecision caused by wide CI and possible publication bias. Certainty in evidence was upgraded due to large effect size.
‡‡Continuous abstinence was defined as abstinence for a period of time rather than at a time point. This definition included both undefined and defined time periods.
§§Moderate certainty in evidence due to imprecision caused by wide CIs and possible public bias but balanced by large effect size.