Treating postpartum affective and/or anxiety disorders in a mother-baby day hospital: preliminary results

Abstract Introduction Women experiencing perinatal mental-health illness have unique needs. The present study analyzes preliminary data about the effectiveness of MBDH in treating postpartum women with affective and anxiety disorders. Methods We analysed 33 mothers with affective and/or anxiety disorders treated at the MBDH with their babies between March 2018 and December 2019. All women were assessed at admission, discharge and three months after discharge. Outcomes included symptoms of depression (EPDS) and anxiety (STAI-S), mother-infant bonding (PBQ) and functional impairment (HoNOs). We also assessed the clinical significance of changes in patients’ scores on these scales and patient satisfaction. Results At discharge, no patients still met the full criteria for the main diagnosis. Between admission and discharge, symptoms of depression and anxiety, mother-infant bonding, functional impairment and autonomy in caring for babies improved significantly. These gains were maintained at three months follow-up. Patient satisfaction was high. Conclusions These preliminary results suggest that multidisciplinary intervention for postpartum women with affective or anxiety disorders at the MBDH improves maternal psychopathology, mother-infant bonding and mothers’ ability to care for their babies. MBDHs are a promising approach for delivering specialised perinatal mental-health care for mother-baby dyads. KEYPOINTS Mother-baby day hospital (MBDH) could be an adequate device for women with perinatal mental disorders Multidisciplinary intervention is effective for treating postpartum anxiety and depressive disorders. Interventions at MBHD improve mother-infant bonding and mothers’ ability to care for their babies. Further research is needed to assess the effectiveness at long term not only on maternal health also on child neurodevelopment.


Introduction
about 20% of women suffer from a mental disorder during gestation and/or after birth (Navarro et al., 2008).Because mothers' mental health also affects their children, perinatal mental disorders are an important public health problem.the most prevalent perinatal mental disorders are affective and anxiety disorders (Borri et al., 2008;Navarro et al., 2008;Vesga-lópez et al., 2008).Moreover, approximately one in three women with postpartum depression suffer from a bonding disorder (Palacios-hernández, 2015) that can affect the mother's responsibility or sensitivity as well as the general care of the baby.
Despite the importance of detecting perinatal mental health problems they are mostly not diagnosed.Webb et al. (2021) conducted a systematic review to identify the barriers and facilitators to implementing assessment, care, referral and treatment for perinatal mental health into health and social care services.the most important were individual (e.g., an inability to attend treatment), health-care professional (e.g., training), interpersonal (e.g., trusting relationships), organisational (e.g., clear referral pathways), political (e.g., funding) and societal factors (e.g., stigma and culture).Recently, the estimated cost of untreated perinatal mood and anxiety disorders from conception to five years postpartum is around $14 billion in the United states alone, with 65% of the cost attributable to mothers and 35% to children (luca et al., 2020).international organisations have pointed out the need to improve the detection, prevention and treatment of perinatal mental disorders and have recommended developing approaches where specialised perinatal mental health professionals can treat mothers and their children together.
since evidence from scientific studies showing the benefits of keeping women experiencing perinatal mental health crises together with their babies began to accumulate in the second half of the twentieth century, specialised mother-baby units have been implemented in hospitals in many countries, although they are most prevalent in the United Kingdom, France and australia (connellan et al., 2017).two systematic reviews found encouraging evidence that joint hospitalisation of mothers and their babies has a positive impact on maternal mental health and mother-infant relationships (connellan et al., 2017;Gillham & Wittkowski, 2015).less evidence is available about the effectiveness of mother-baby day hospitals (MBDh), a partial hospital model of care in which a multidisciplinary team comprising perinatal psychiatrists, perinatal adult and child clinical psychologists, specialised mental health and paediatrics nurses, and social workers provides intensive services during the day to patients whose symptoms and/or functional impairment require a higher level of care than traditional outpatient services.
there were no specific mental health programs offering integrated intensive perinatal care to women in spain until 2018, when a tertiary university hospital in Barcelona established its MBDh in line with the international clinical Guidelines on Perinatal Mental health (austin et al., 2017;National institute for health and clinical excellence, 2014;scottish intercollegiate Guidelines Network, 2012).the current study analysed preliminary data on the effectiveness of intensive multidisciplinary treatment at this MBDh in mothers with a depressive and/or anxiety disorder in the first year postpartum on mothers' health, their ability to care for their infants and mother-infant bonding.

Design
this prospective study compared variables related to mothers' health and ability to function before and after treatment at the MBDh.

Participants and procedures
the study population included all women diagnosed with mood or anxiety disorders during the first year postpartum according to Diagnostic and statistical Manual of Mental Disorders-Fourth edition (DsM-iV-tR) who were admitted to and discharged from the MBDh between March 2018 and December 2019.the only other inclusion criterion was the ability to read and respond to self-reported measures.the institutional review board approved the study (hcB/2020/1360).
a perinatal psychiatrist diagnosed the acute episode, evaluated medical and psychiatric comorbidities and assessed the risks and benefits of psychopharmacological treatment.all patients attended cognitive-behavioral group therapy for anxiety and depression, a mother-infant intervention group and nursing psychoeducation groups.every week, all participants attended at least two one-day sessions consisting of a psychology group, a nursing group and, if necessary, a psychiatric visit.Patients were discharged when their symptoms were under control, and they were considered able to function in daily life.
the MBDh clinical staff (psychiatrist, psychologist and nurse) assessed patients at admission, at discharge and three months follow-up.
at admission and discharge from the MBDh, a clinician used a scale developed by our team to assess maternal autonomy for basic care behaviours.For the present study, mothers' behaviour was classified as autonomous (not requiring assistance) or not autonomous (requiring assistance).
at discharge and three-month follow-up, all women were assessed (liFe, ePDs, stai-s, PBQ, hONOs) and, at discharge, they also asked to complete an anonymous questionnaire to assess patients' satisfaction and the perceived usefulness of the treatment.
Online supplement 1 provides detailed descriptions of the instruments used.

Statistical analysis
Descriptive statistics were compiled for all variables.to evaluate the effectiveness of the program, we used paired t-tests, and we calculated effect sizes with cohen's d.Moreover, we calculated the clinical significance for the primary outcome measures (ePDs, stai and PBQ) based on Jacobson and truax (1991) classification of meaningful change.this system uses two criteria, cut-off (post-test score within a functional distribution of scores) and Reliable change index (Rci) (to ensure change is not due to measurement errors), to classify subjects as recovered (fulfilling both criteria), improved (fulfilling the Rci but not the cut-off ), unchanged (fulfilling neither criteria) or deteriorated (fulfilling the Rci in a worsening direction).We used sPss 25.0 for all analyses.
the baby was the mother's first child in 22 (66.7%)dyads, the pregnancy had been planned in 24 (72.7%), and breastfeeding was underway in 23 (71.9%).a total of 25 (75.8%)women reported a prior psychiatric disorder, including 3 (9.1%)that required hospitalisation (2 during the perinatal period).a total of 20 (60.6%) women reported a family history of psychiatric disorders, including 2 (6.1%) that reported a family history of suicide.
On the MiNi, 25 (75.7%)women met DsM-iV-tR criteria for a major depressive episode and 8 (24.2%) for anxiety disorder (panic disorder with or without agoraphobia).the MiNi identified suicidal ideation in 18 (54.6%)women; precipitating suicidal ideation was the most common type, identified in 8 (24.2%) women.Responses to ePDs-item 10 confirmed thoughts of harming in 18 women.at least one comorbidity was present in 18 (54.6%)women (obsessive compulsive disorder and generalised anxiety disorder were the most common).On the Pychiatric status Rating (PsR) yielded by the liFe, all women met the criteria for a disorder (PsR ≥ 5) and 4 (12.1%) had extreme functional impairment (PsR = 6).
all 33 mother-infant dyads completed the treatment at the MBDh, attending a mean of 45 ± 20.6 sessions; 32 (97%) were also assessed at three months follow-up.all women received antidepressant treatment (mostly sertraline, followed by venlafaxine) and 21 (63.6%) with high doses.seven mothers (21.2%) received combined treatment (5 antidepressant plus antipsychotic, 2 antidepressant combination).63.6% of women (n = 21) also required treatment with low-to-medium doses of benzodiazepines at some time of intervention.

Clinical improvement and clinical significance
table 1 reports the ePDs, stai-s, PBQ and hoNOs scores at admission, discharge and three-month follow-up as well as mean differences and effect sizes on each instrument.all scores improved significantly between admission and discharge; the corresponding effect sizes were large.all scores also improved significantly between discharge and three-month follow-up; the corresponding effect sizes were small or medium.
table 2 shows the clinical significance of changes in depression, anxiety and mother-infant bonding, reporting the percentage of patients allocated to each intervention outcome category according to Jacobson and truax (1991) classification of meaningful change.although none of the patients' conditions had worsened after treatment, ePDs, stai-s and PBQ remained unchanged in five (15.2%) patients with scores in the dysfunctional range.PBQ scores also remained unchanged in nine (27%) women with scores in the functional range.From 6% to 15% of the women improved at discharge from the MBDh (15% according to the ePDs, 12% according to the stai-s and 6% according to the PBQ).More than half of the women had achieved recovery at discharge from the MBDh (70% according to the ePDs, 57% according to the stai-s and 52% according to the PBQ).at three months follow-up, recovery rates were higher according to all scores, especially according to the ePDs (n = 26; 78%).

Autonomy in maternal care behaviours
at admission to the MBDh, only four (12.1%) mothers were able to take care of their babies without support.at discharge, 24 (72.7%)mothers were self-sufficient in caring for their babies (p < .001);at three months follow-up, all these women remained self-sufficient.statistically significant differences were observed in all domains, except in changing diapers (p = .063)and putting the baby to sleep (p = .250)(Figure 1).

Satisfaction
all women completed the anonymous patient satisfaction questionnaire at discharge.Mean ratings for all aspects of the MBDh were >9 on the scale ranging from 0 to 10. all women stated that they would recommend the program to others.Figure 2 shows women's satisfaction with different aspects of the MBDh and perceived usefulness of the program at discharge.

Discussion
Our analyses show that maternal symptoms of depression and anxiety, mother-baby bonding and mothers' autonomy in caring for their babies had improved significantly at discharge from the MBDh and that these improvements were maintained at the three-month follow-up.Our results add to the growing body of contradictory evidence from studies conducted at MBDhs.Preliminary results of an MBDh for pregnant and postpartum woman in the United sates (howard et al., 2006) found that 53% of the patients were classified as recovered and 25% as improved.the lower rates in recovery in their study could be explained, in part, by the brevity of their program, with an average stay around seven days.a more recent study (Kim et al., 2021) reported significant improvements (p < .001) in self-report scales assessing depression, anxiety and maternal functioning, although these authors did not evaluate the meaningfulness of clinical change reflected in these results.
although the MBDh uses a partial hospitalisation approach, the severity of our patients' disorders allow our results to be compared with those of inpatient mother-baby units.Our sample comprised women with moderate-severe affective/anxiety disorders: half had comorbidities and/or suicidal ideation at admission, three-quarters had a prior history of mental illness and 9% had been admitted to psychiatric hospitals.all required combined psychological and psychopharmacological treatment and 15.2% required two or more drugs (other than low-dose benzodiazepines).Moreover, our patients' mean scores at admission are similar to those in reports from inpatient mother-baby units: ePDs (christl et al., 2015;Gilden et al., 2020), PBQ (Gilden et al., 2020) and hoNOs (stephenson et al., 2018).
as in our study in the MBDh, studies done at inpatient mother-baby units report significant improvement in depressive and anxiety symptoms at discharge (christl et al., 2015;Meltzer-Brody et al., 2014).Our results for mental health and well-being as measured by the hoNOs are also similar to those reported in studies of inpatient mothers with any mental health disorder (Wright, stevens, et al., 2018) and those with a primary diagnosis of depression or anxiety (stephenson et al., 2018).importantly, none of the women in our study still met the full criteria for the main diagnosis (PsR ≥ 5) at discharge, and the improvement in depressive and anxiety symptoms was clinically meaningful (70%-85% of women were classified as achieving clinically significant recovery or improvement, depending on the score used).these findings are similar to those reported by christl  2015) in mothers treated for postpartum anxiety or depressive disorder in an inpatient mother-baby unit, where 73.3% of women recovered from their symptoms.similarly, studies reporting audits of inpatient mother-baby units in the United Kingdom and France that assessed psychopathological improvement with the Marcé clinical checklist (appleby & Friedman, 1996) reported around 70% of patients were 'symptom-free or considerably improved' at discharge (Glangeaud-Freudenthal, 2004;Glangeaud-Freudenthal et al., 2011;salmon et al., 2004).We found that treatment at the MBDh not only improved maternal outcomes, but also mother-infant bonding, as evidenced by statistically and clinically significant reductions in PBQ scores.these findings are consistent with those reported in a systematic review of studies assessing mother-infant relationships with self-reported scales, staff-rated scales or observational measures (Gillham & Wittkowski, 2015).Nevertheless, although mean PBQ scores at discharge were above the cut-off of 13 for bonding disorders (torres-Giménez et al., 2021), 15% of our patients had scores within the dysfunctional range.these results are in line with those reported by Gilden et al. (2020), who found improved bonding overall but impaired bonding at discharge in 18% of women with postpartum depression.
in our study, improvements in mother-baby bonding and anxiety symptoms were stable between discharge and follow-up, although scores remained in the moderate clinical range for over a third of women.these results differ from those reported by Reilly et al. (2019), in which self-reported mother-infant attachment declined between discharge and follow-up in 93.3% of women and symptoms of depression and stress increased in 34.6%.Depressive symptoms are negatively associated with the  quality of (tichelman et al., 2019), and taken together, these results highlight the importance of dyad-centred perinatal mental health services after discharge.Patients were discharged from the MBDh after showing improvements in their symptoms and functioning in daily life, including the ability to care for their baby.Our results show that mothers with anxiety and depressive disorders can have difficulties in caring for their babies and that the MBDh can help them overcome these difficulties.Whereas only 12.1% of mothers were self-sufficient in caring for their babies without support at admission, at discharge most mothers had become autonomous in a wide range of baby-care behaviours.thus, in postpartum mental illness, it is essential to evaluate mothers' ability to care for the baby and offer appropriate interventions to help improve their parenting, including perinatal mental health services (salmon et al., 2004; stephenson et al., 2018).
the improvements in symptoms, functioning and mother-infant bonding were sustained during the three months after discharge from the MBDh, but depressive symptoms in the dysfunctional range persisted in approximately 10% of women.these long-term outcomes are better than those reported after discharge from inpatient mother-baby units.Wright et al. (2020) reported that maternal psychopathology and the mother-infant relationship deteriorated significantly in around one-third of women in the three months after discharge from a mother-baby unit.Reilly et al. (2019) found that during the 15 weeks after discharge from a mother-baby unit symptoms of depression worsened in 63% of mothers and mother-infant bonding deteriorated in 93.3%.these differences might be due to difficulties in adjustment after discharge from inpatient mother-baby units, where women returning home face parenting, household and occupational demands with decreased support and consequent increase in perceived stress (connerty et al., 2016).By contrast, women being treated at MBDhs remain connected to their home environment and responsibilities, and this partial hospitalisation model probably eases transitioning after discharge.additionally, all patients are followed up by a psychiatrist in an outpatient clinic managed by the same department until 12 months postpartum or transfer to another outpatient mental health service.these measures ensure continuity of care, facilitating adaptation to outpatient monitoring and rapid detection of signs of relapse or risk situations such as discontinuation of treatment.
Mother-baby dyads attended a mean of 45 sessions in the MBDh.each session represents a single day in the MBDh in which the patient attended at least one psychological intervention group and one nursing group.all patients attended at least two days a week; those with more severe symptoms initially attended more frequently.this model is difficult to compare with inpatient mother-baby units, where mean length of stay ranges from 7 days to 12 weeks and differs depending on the unit's nature and purpose (connellan et al., 2017).the most valid comparison would probably be with stephenson et al. 's (2018) study of women admitted to an inpatient mother-baby unit in the United Kingdom, where the mean length of stay was 75 days for those admitted for depression and 60 days for those admitted with anxiety/ obsessive-compulsive disorder.
Women attended at the MBDh were highly satisfied with all aspects of the program, and they considered it useful.these results are consistent with previous research in inpatient mother-baby units, where women's experiences were positive overall (connellan et al., 2017).Women generally prefer specialised care focused on perinatal women to more generic care in general psychiatric wards.Women value having their babies with them and consider that it helps their recovery as well as the support they receive in parenting and practical childcare (Griffiths et al., 2019;Neil et al., 2006;Wright, Jowsey, et al., 2018).
to our knowledge, this is the first study to investigate the effectiveness of an MBDh for mothers with affective and/or anxiety disorders that analyses not only psychiatric outcomes but also mother-infant bonding and maternal care behaviours over a three-month period.Our preliminary results are encouraging, showing significant improvement and large effect sizes in all areas evaluated.the major limitations of the study are the small sample and the lack of a control group.Further studies including more patients, especially randomised clinical trials, would enable stronger conclusions about the effectiveness of this approach.however, randomising women with perinatal mental disorders poses some practical and ethical problems, partly due to the lack of specialised perinatal mental health services in our country.another limitation is that specific questionnaires for comorbid diagnoses (e.g., OcD) have not been measured as this was not a primary objective of the study.

Conclusions
this analysis of the only MBDh program in spain suggest that this approach of delivering, intensive, multidisciplinary, specialised perinatal care to with affective and/or anxiety disorders together with their babies is feasible and effective.

Table 1 .
comparison of scores at different timepoints in women admitted to a mother-baby day hospital.