Short-term outcome of mothers with severe mental illness admitted to a mother baby unit.

BACKGROUND: Mother baby psychiatry units (MBUs) are the expected standard of inpatient care internationally for postpartum mothers with severe mental illness (SMI) and favorable outcomes for mother infant dyads have been reported from these services. However, there are very few such units in low- and middle-income countries. The current study aimed to assess the short-term outcome of mothers in SMI admitted to an MBU in India.
MATERIALS AND METHODS: Mother infant dyads admitted over a year in the MBU were assessed in detail at admission, discharge, and at 3 months. Tools used included the Brief Psychiatric Rating Scale (BPRS), Young's Mania Rating Scale, Edinburgh Postnatal Depression Scale, and Clinical Global Impression. Mother infant interaction was assessed using the NIMHANS maternal behavior scale.
RESULTS: The mean age of the 43 mothers admitted in this period was 27.3 ± 6.2 years. For 27 (62.8%) mothers, this was a first episode of psychosis. Affective disorders and acute psychotic disorders were the most common diagnoses. The average duration of stay in the MBU was 25 days. While all mothers showed significant improvement at discharge, mothers with a first episode in the postpartum had higher BPRS scores (Ws = 309.5, P = 0.02) as compared to the others. At 3 months following discharge, all mothers sustained the improvement achieved. DISCUSSION: Mothers with SMI admitted to an MBU showed significant clinical and dyadic improvement at discharge which was sustained at 3 months. However, the lack of a control group limits the generalizability of the current findings.
CONCLUSION: The study highlighted a favorable short-term outcome among mothers with SMI admitted to a MBU facility. Copyright:

INTRODUCTION

Postpartum period is considered to be a vulnerable phase for new-onset psychiatric illness or exacerbation of preexisting psychiatric illness.[1] Severe mental illness (SMI) which includes bipolar affective disorders, schizophrenia, and related psychotic disorders during postpartum is considered as a psychiatric emergency in view of its significant risk to the mother and her infant. Further, an early separation of infant from an unwell mother can have negative consequences such as aggression and negativity during childhood[2] as well as delay the recovery among postpartum mothers with mental illness.[3]

Mother baby unit (MBU) is a psychiatric inpatient setup that provides comprehensive care to women with mental illness along with her baby during the postpartum period. The comprehensive care involves a focus on mother-infant interactions and the child's development in addition to clinical care of the mother. Currently, most of the MBUs are located in high-income countries (HICs) with sparse services in low- and middle-income (LAMI) countries.[4] The available data have shown good outcome among mothers with postpartum mental illness admitted along with their infant in an MBU setting in HIC.[5]

A systematic review looking at the outcome of mothers admitted to psychiatric MBUs across the world included 10 studies out of which 9 were from HIC and only one study from LAMI country. The review found that 69% to 78% of mothers improved in their symptoms at the time of discharge from the MBU.[5] The factors responsible for poor outcome among mothers included diagnosis of schizophrenia, personality disorders, poor social integration, severe behavioral disturbance, low social class, and poor relationship or psychiatric illness in the partner.[6] Specific factors responsible for poor mother infant outcomes included, mothers with schizophrenia and other psychotic disorders with lack of mothering skills,[7] mothers with postnatal depression who had poor bonding with their babies,[8] young mothers with higher psychosocial risk with low parenting confidence,[9] mothers with both psychotic and nonpsychotic depression with feelings of insecurity with their infants compared to mothers with postpartum mania.[10] The review also highlighted that at the time of discharge physical harm to baby was seen more with affective disorders[11] and infant being separated if mother had psychosis.[1112] There are limited MBUs located in LAMI countries, and the only available data are from MBU situated in India. The outcome of postpartum mothers from India reported that 80% had improved completely while rest had residual symptoms at discharge.[4] The observational study showed that multidisciplinary MBU care significantly improved the psychopathology and mother-infant interaction among women with varied diagnosis.[13] However, the authors were not clear whether such improvement in mother infant interactions were maintained postdischarge.[13] The aim of the current study was to investigate the short-term clinical outcome and its association with demographic and clinical factors that might influence the outcome among mother infant dyads admitted to an MBU in India.

MATERIALS AND METHODS

This prospective study included mothers who were admitted to the Mother Baby Perinatal Psychiatry Unit at National Institute of Mental Health and Neuro Sciences (NIMHANS), Bengaluru during a 1-year period (March 2016 to February 2017). The study included mothers with the age range of 18–35 years, admitted with their infants to the MBU with a diagnosis of postpartum SMI. Mothers with history of intellectual disability and personality disorders were excluded. The study was approved by the Institutional Ethics Committee. Mothers participated in the study after providing written informed consent.

Sociodemographic and clinical details such as past and family history, duration of postpartum onset, diagnosis, duration of hospital stay, administration of electro convulsive therapy (ECT), and psychosocial risk factors[914](history of previous abortion, significant grief in past 1 year, dissatisfaction about the gender of the baby, traumatic childbirth, childhood trauma, and unwanted pregnancy) were collected using a semi-structured pro forma. Maternal psychopathology was rated using Brief Psychiatric Rating Scale (BPRS),[15] Young's Mania Rating Scale (YMRS),[16] Edinburgh postnatal depression scale (EPDS),[17] and Clinical Global Impression (CGI)[18] Maternal behavior toward her infant was recorded using NIMHANS Maternal Behavior Scale (NIMBUS).[19]

NIMHANS maternal behavior scale (NIMBUS)

The NIMBUS is an objective assessment of mother infant interactions among mothers with SMI. It has five domains and 1 item on reasons for separation from infant during psychiatric illness. Each mother-infant dyad was observed in terms of care for the baby's basic needs, affectionate behavior, significant incidents, overall assessment of safety, how does the mother handle separation from the baby and whether the mother is separated from baby last few days and reasons for the same.

The above assessments were done at baseline and at the time of discharge.

The short-term outcomes were assessed after 3 months from the time of their discharge from the hospital. Short-term outcome assessments included CGI-improvement and NIMBUS. Telephonic follow-up was done if the patient failed to visit the hospital by 3 months following the discharge.

Statistical analysis

Statistical analysis was performed using the SPSS Inc. Released 2007. SPSS for Windows, Version 16.0. Chicago, SPSS Inc. version 16. Descriptive statistics such as frequency and mean were used to describe the sociodemographic variables and total scores of the scales. Wilcoxon signed-rank test used for comparing mean scores of BPRS, YMRS, and EPDS assessed at three points of time. Friedman's test was used for comparing scores of CGI and maternal behaviors rating scales at three points of assessment.

RESULTS

During the study period, 43 mother-infant dyads with mothers having postpartum SMI who met inclusion criteria were recruited. The mean age of the mothers in years was 27.3 (6.2) years. Majority (90.7%) belonged to Hindu religion, having secondary (8–10 years) level of education (44.2%) and from rural background (60.5%). Except one woman all others were married. Most of the mothers (86%) were home makers.

Clinical details

First episode of SMI during the postpartum period was seen in 27 (62.8%) mothers. Both past and family history of mental illness was seen in 21 (49%) mothers. The mean duration of onset of symptoms in the postpartum period was 11.3 (10.1) weeks (range: 1st week–36th weeks). None of the mothers had catatonia at the time of evaluation. Ten (23%) mothers had a history of deliberate self-harm attempts. The most common psychiatric diagnosis was affective disorders (bipolar disorder and severe depression with psychotic symptoms) (40%) followed by acute psychosis (39%), schizophrenia (14%), and schizoaffective disorders (7%). Mean duration of their hospital stay was 21.5 (14.2) days (range: 7 days to 94 days). More than half the mothers (53.49%) received Electro Convulsive Therapy (ECT) during inpatient care. Majority (81%) of the mothers perceived that they had adequate support system and 10 (23%) mothers had a history of family violence. Detailed evaluation during inpatient care noted comorbid personality traits in 9 (21%) mothers. Other psychosocial risk factors noted were history of previous abortion in 6 (14%), significant grief and dissatisfaction about the gender of the baby in 4 (9%) and traumatic childbirth, childhood trauma, and unwanted pregnancy in 1 (2%) mother each. All mother-infant dyads received individualized multidisciplinary care.

Clinical outcome at admission and discharge

The median scores on BPRS, YMRS, and EPDS at admission and discharge are shown in Table 1. The severity of illness was recorded at admission and discharge. Wilcoxon Signed-rank test was used to compare the total score at admission and discharge. Significant reduction in the total score was observed in all the parameters (P < 0.001) [Table 1]. The BPRS at discharge among those who had the 1st episode postpartum illness was significantly higher than those with worsening of preexisting psychiatric illness in the current postpartum (Ws = 309.5, P = 0.02).

Table 1

Comparison of Rating Scales at admission and discharge

Scale	Time of assessment	Median	IQR	Z	P
BPRS	Admission	69.50	19	−5.646	<0.001
	Discharge	32.0	12.5
YMRS	Admission	25	22	−5.225	<0.001
	Discharge	10	6.25
EPDS	Admission	18	5	−5.012	<0.001
	Discharge	8	6.75

IQR – Interquartile range; BPRS – Brief Psychiatry Rating Scale; YMRS – Young’s Mania Rating Scale; EPDS – Edinburgh Postnatal Depression Scale

Clinical Global Impression (CGI) and NIMHANS Maternal Behavior Scale (NIMBUS) at admission, discharge, and follow-up

Clinical Global Impression (CGI) and NIMHANS Maternal Behavior Scale (NIMBUS) at admission, discharge and follow-up. The total score on CGI and individual item score on NIMBUS was recorded at admission, discharge, and at 3 month follow-up. Friedman test was used to compare the total scores on CGI and individual item score on NIMBUS. Significant improvement was noted at discharge and it was sustained over a period of 3 months follow-up (P < 0.001) [Table 2].

Table 2

Comparison of clinical global impression and National Institute of Mental Health and Neuro Sciences Maternal Behavior Scale at admission, discharge, and follow-up

Scale	Time of assessment	Median (mean)	IQR (SD)	t	P
CGI	Admission	5 (4.95)	0 (0.216)	79.480	<0.001
	Discharge	2 (2.093)	0 (0.610)
	Follow-up	2 (2)	0 (0)
NIMBUS
Care	Admission	8 (8.33)	7 (3.400)	54.642	<0.001
	Discharge	4 (4.744)	0 (1.852)
	Follow-up	4 (4)	0 (0)
Affection	Admission	10.4 (11.40)	10 (6.233)	51.283	<0.001
	Discharge	5 (5.691)	0 (3.573)
	Follow-up	5 (5)	0 (0)
Significant incidents	Admission	5 (6.05)	1 (1.812)	21.593	<0.001
	Discharge	5 (4.953)	0 (1.233)
	Follow-up	5 (5)	0 (0)
Overall safety	Admission	26 (27.53)	16.0 (9.493)	58.403	<0.001
	Discharge	15.30 (17.535)	1.0 (6.370)
	Follow-up	15 (15)	0 (0)
Handling separation	Admission	2 (2.53)	3 (1.30)	49.523	<0.001
	Discharge	1 (1.256)	0 (0.276)
	Follow-up	1 (1)	0 (0)
Total bonding score	Admission	32 (32.60)	17 (9.965)	65.871	<0.001
	Discharge	17 (19.71)	2 (6.653)
	Follow-up	17 (17)	0 (0)

CGI – Clinical global impression; NIMBUS – NIMHANS Maternal Behavior Scale; NIMHANS – National Institute of Mental Health and Neuro Sciences; IQR – Interquartile range; SD – Standard deviation

Factors influencing the clinical outcome

Fisher's exact test was used to evaluate association between outcome and clinical variables (family and past history of mental illness and/or suicide attempts). No such association could be made. In view of small sample size, complex statistical analysis was not done.

Correlations between clinical outcome and mother-infant attachment

There were no significant correlations between clinical outcome and mother-infant bonding (objective assessment) except scores of YMRS and total bonding scores along with their subscales though strength of association was low (correlation coefficient – 0.522) using nonparametric spearman rho correlation.

DISCUSSION

The current study aimed to assess the clinical outcome of mothers with postpartum SMI admitted along with their infants in an MBU from a low resource setting. Forty-three mothers who were evaluated at admission, discharge and at 3-month postdischarge had significant improvement in their psychotic, depressive, and manic symptoms at discharge and sustained the improvement over the 3-month period. Most of the previously published literature on maternal outcome[6791320] have reported significant improvement at discharge. One of the largest studies in the UK evaluated 1081 mothers admitted to 8 MBUs over a 6-year period reported that 78% mothers had good outcome with respect to clinical improvement and 80% improvement in parenting outcome. The joint admission of mother infant dyad has been found to improve the outcome of mothers not only in their psychopathology but also in terms of improvement in their parenting skills.[1320] A similar finding of 69% of the mothers showing improvement in clinical symptoms in 869 women admitted to 13 MBUs in France[6] and Australia.[79] Our study did not show any specific factor that could have has been reported influenced the outcome except the fact that women with the 1stepisode postpartum illness had higher scores on BPRS as compared to women who had past psychiatric illness. It is possible that around 79% of our sample had a diagnosis of mood disorder and acute psychosis which are known to have favorable outcomes in previous studies in comparison to schizophrenia.[67] However, higher scores on BPRS at discharge among the 1stpostpartum episode in our sample could be because of the higher prevalence of mood incongruent psychotic symptoms in the 1stonset psychosis occurring in postpartum.[21] There was no correlation with maternal psychopathology and mother infant bonding. Similar findings have been documented in previously published literature with improvement in maternal sensitivity at the time of discharge.[13] Only a weak association was noted between YMRS scores of mothers with mother-infant bonding. Clinical status of the mothers might not be directly associated with parameters of mother infant relationship.[13]

In our study, the improvement in the clinical outcome and mother infant bonding sustained over a period of 3 months. It is quite possible that all mother infant dyad had received individualized care in addition to standard care during inpatient admission. Other possible reasons could be over representation of mood disorders and acute psychosis, perceived social support. However, there is limited literature on short-term outcome on postpartum SMI form low resource settings. A review on long-term outcome of postpartum psychosis has shown that 43.5% of women did not have episodes outside the perinatal period.[22] Apart from that our center has good after care facility in terms of initial early follow-up and providing helpline number to contact in case of difficulties.[23]

The strength of the study include that, this was first prospective short-term outcome study involving mother-infant dyads admitted to an MBU located in LAMI country. All mothers underwent structured and validated assessments at three different point in time (admission, discharge, and 3-month follow-up). The researcher was able to follow-up all the mothers who consented for the study at 3 months postdischarge. This could be because the researcher had good rapport with all the mothers recruited in the study. This can be strength as well as weakness of the study because in real life scenario there will be dropouts in a prospective study design. This itself can have impact on the study results.

The major limitation in this study is the lack of a control group, and hence, it is difficult to infer that the improvement among mothers was due to the MBU care. Another limitation was small sample size and inability to do complex analyses to understand the role of predictors of outcome. Psychosocial risk factors were assessed through the open-ended questions to mothers rather than structured interview. All mother infant dyads in MBU received standard care with additional interventions based on the felt need. We did not record what specific interventions mother received during the study period. All the assessments were done by the same investigator at three points of time which could have biased the observations.

CONCLUSION

The study highlights favorable outcome among mothers admitted to a MBU facility. Lack of control group limits inferences in this study. Future research should focus on long-term follow-ups of mother infant dyads, cost-effectiveness of MBU facility.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.