Postnatal depression (PND) affects up to one in four mothers. However, they may experience barriers to access to conventional treatments, indicating a need for alternatives such as arts-based interventions. A previous trial showed that a 10-week singing intervention could alleviate symptoms of PND.

To evaluate, in a larger sample and across a longer timeframe than previously, the clinical effectiveness, implementation effectiveness and cost-effectiveness of the Melodies for Mums (M4M) singing intervention for symptoms of PND.

One-hundred and ninety-nine mothers experiencing symptoms of PND (Edinburgh Postnatal Depression Scale score ≥10) and their babies were randomised to 10 weeks of in-person singing sessions (M4M, n = 133) or an active control (existing community-based mother–baby activities, n = 66). Mothers were re-assessed at weeks 6, 10, 20 and 36 for depression, healthcare use for themselves and their babies, and health-related quality of life according to the EQ5D-3. The perceived acceptability (Acceptability of Intervention Measure), appropriateness (Intervention Appropriateness Measure) and feasibility (Feasibility of Intervention Measure) of the activity were also assessed at week 6. Trial registration number: NCT04834622.

Mothers in both groups experienced attenuation of depressive symptoms by week 10; however, those in the singing group maintained lower EPDS scores than those in the control group at week 20 (10.7 v. 12.2 (mean difference 95% CI [−2.96, −0.22]), P = 0.023) and week 36 (9.85 v. 11.4 [−2.93, −0.19], P = 0.026). Mothers in the singing group were also more likely to remain in the study (77 v. 57%, χ2(1) = 12.92, P < 0.001) and found their programme more acceptable (4.75 v. 4.0 [0.25, 0.83], U = 2436.5, P < 0.001), appropriate (4.25 v. 3.88 [0.12, 0.62], U = 2241.5, P < 0.001) and feasible (4.75 v. 4.0 [0.41, 0.91], U = 2568.0, P < 0.001). Finally, M4M was associated with 15 extra days of health and was found to be cost-effective (£126–539 per dyad).

M4M had a long-lasting effect on symptoms of PND and was perceived to be more suitable than existing activities; thus, M4M represents a worthwhile investment for healthcare systems as an intervention for mothers experiencing symptoms of PND.

Persons with behavioral variant frontotemporal dementia (bvFTD) have been shown to exhibit altered morality, manifested as atypical utilitarian tendencies towards sacrificial moral dilemmas. This takes the form of endorsing harmful actions towards single individuals, including vulnerable or relationally close individuals (e.g. children, loved ones), in order to promote the greater good for the community or society as a whole. The dual process model of moral cognition interprets such tendencies as deriving from a lack of emotional engagement, whereas moral emotion theory views them as selective impairment in prosocial sentiments. We hypothesized that both the widespread neuropsychological practice of using sacrificial moral dilemmas to evaluate moral reasoning, and these tests' overreliance on quantitative results, inadequately represent how persons with bvFTD reason and feel while responding to moral dilemmas. To evaluate this hypothesis, we applied a mixed-methods approach to identify the reasoning, motivations, and emotional experiences of bvFTD persons during their deliberation about moral scenarios.

We conducted semi-structured interviews with 14 participants: 7 persons with bvFTD & 7 older healthy controls. Transcripts were coded in Atlas 5.0 to characterize the underlying reasoning, emotions, response processes, and values that emerged when responding to a structured set of moral dilemmas. Our dilemmas measured utilitarian reasoning holistically by incorporating both sacrificial and impartial/altruistic components, as suggested by the 2-dimensional model of utilitarianism.

Unexpectedly, bvFTD persons articulated a prosocial compass when asked about their values, stating they were organizing their choices predominantly around kindness and altruism, even when they were making choices to harm loved ones or vulnerable individuals to promote the greater good. During moral deliberation, persons with bvFTD showed significantly less metacognition (bvFTD = 10%, HC = 90%) but reported more positive emotions (joy; bvFTD = 83%, HC = 17%) than negative (frustration; bvFTD = 30 %, HC = 70 %) compared to controls. Qualitatively, this observed emotional outlook was typically coupled with a more rigid, simplistic viewpoint (e.g., "I felt great, it was a no brainer"), suggesting a moral understanding lacking emotional nuance and complexity.

Our data showed that bvFTD persons' utilitarian responses to moral dilemmas did not arise from an emotionally flat or antisocial cognitive perspective, but instead were guided by positive emotionality, simplistic reasoning, and prosocial values. These findings challenge the current understanding of the reasoning processes and experiences of persons with bvFTD and highlight the importance of incorporating mixed method approaches in dementia research that take into consideration the viewpoint of the cognitively compromised individual.