Attention and social communication skills of very preterm infants after training attention control: Bayesian analyses of a feasibility study

Abstract

Background: Very preterm (VP) infants (born 28 to <32 weeks of gestation) are at risk of cognitive delays and lower educational attainments. These risks are linked to anomalies in attention and information processing that emerge in the first years of life. Early interventions targeting attention functioning may equip VP infants with key building blocks for later attainments.

Methods: We tested the feasibility of a randomised trial where VP infants took part in a computerised cognitive procedure to train attention control. Ten healthy VP infants aged approximately 12 months (corrected age) and randomly allocated with 1:1 ratio to the training (interactive computerised presentations) or an active control procedure completed the study. Before and after the training programme, participating infants completed a battery of screen-based attention tests, naturalistic attention and communication tasks, and temperament assessments. In a previous study we analysed the data concerning feasibility (e.g. recruitment and retention). In the paper presented here we considered the infants' performance and used Bayesian regression in order to provide credible treatment estimates considering the data collected.

Results: Estimates indicate moderate treatment effects in visual memory: compared to controls, trained infants displayed improvements equivalent to 0.59 SD units. Trained infants also improved in their abilities to attend to less salient stimuli presentations by 0.82 SD units, compared to controls. However, results did not indicate relevant gains in attention habituation or disengagement. We also reported moderate improvements in focused attention during naturalistic tasks, and in directing other people's attention to shared objects.

Discussion: The results warrant further investigation concerning the effectiveness of training attention control in VP infants, extending this line of research beyond our small and homogeneous sample of healthy VP infants. This study also emphasises the utility of Bayesian approaches in estimating potentially relevant effects in small samples or exploratory studies. The scope for further research on early attention control training is discussed in light of studies indicating VP children's susceptibility to positive environmental inputs.

Trial registration: Registration ID: NCT03896490. Retrospectively registered at Clinical Trials Protocol Registration and Results System (clinicaltrials.gov).

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1. CONSORT diagram.

Fig 2. Fitted (i.e.: Estimated) d scores by treatment and age in models that suggested relevant interactions treatment by age.

Lines represent estimated scores based on the posterior distribution. The shaded areas represent 89% uncertainty intervals of the scores estimated from the posterior distrition. Dots represent the observe treated and control scores by age.

Fig 3. Estimated distributions of d^T scores (i.e. differences in d scores between treated and controls) by screen-based outcomes.

The dotted lines highlight a d^T = 0, i.e. no difference in performance between treated and controls. Scores on the right side of the dotted lines indicate improved post-test perfromance of the treated compared to the controls. NOTE: These scores were estimated by sampling n = 10,000 observations from the posterior distribution of parameters estimated in each selected model (see Table 5). Where the effect of the treatment was moderated by age, scores for treated and controls were estimated by considering scores at age 13 months.

Fig 4. Estimated distributions of d^T scores (i.e. differences in d scores between treated and controls) by naturalsitic tasks outcomes.

The dotted lines highlight a d^T = 0, i.e. no difference in performance between treated and controls. Scores on the right side of the dotted lines indicate improved post-test perfromance of the treated compared to the controls. NOTE: These scores were estimated by sampling n = 10,000 observations from the posterior distribution of parameters estimated in each selected model (see Table 5). Where the effect of the treatment was moderated by age, scores for treated and controls were estimated by considering scores at age 13 months.