Abstract
Background:
Explanatory models (EMs) influence decision-making related to treatment and compliance. There is little knowledge about belief systems related to postpartum psychosis in different cultures.
Aims:
To study EMs of illness among women with postpartum psychosis and their family members in India.
Method:
A total of 123 women with postpartum psychosis and their caregivers were assessed to understand their beliefs regarding causes of postpartum psychosis using Short Explanatory Model Interview (SEMI). Local names were listed and emerging themes were analysed.
Results:
Respondents often held more than one EM and only one-third held a biomedical EM. Other common models included stressors unique to childbirth, marital stress and supernatural causes. Local names reflected the underlying models.
Conclusion:
Non-biomedical EMs are common in women with postpartum psychosis. Cultural and social factors unique to childbirth appear to influence these models. There is a need to enhance awareness and knowledge about this serious disorder in the community.
Introduction
Pregnancy and childbirth involves a complex social interaction influenced by cultural undercurrents. The emphasis by family and community on childbirth traditions and spiritual beliefs influences a mother’s view about pregnancy, motherhood and associated risks (Kaphle, Hancock, & Newman, 2013). Postpartum period is a vulnerable time for worsening or new occurrence of mental health problems affecting the mother’s quality of life and mother–infant bonding. Severe mental illness during postpartum period described as ‘postpartum psychosis’ can present as a manic episode, severe depression or acute psychosis. Postpartum mental illness carries a high risk of suicide and infant harm, and the condition warrants immediate clinical attention (Chandra, Bhargavaraman, Raghunandan, & Shaligram, 2006; Oates, 2003).
Explanatory models (EMs) for a health problem comprise ideas about cause, course and treatment options for relief from the symptoms (Kleinman, 1980). EMs influenced by culture play an important role in recognition, labelling and help-seeking behaviour for affected individuals and family members (Ryan, 1998). Explanations based on socio-cultural belief systems outside the purview of an individual are likely to be accepted with lesser stigma within the community. Attribution of causes of mental illnesses to supernatural factors, curse or karma and other non-medical factors often lead to help seeking from places of worship and traditional healers (Campion & Bhugra, 1997; Kurihara, Kato, Reverger, & Tirta, 2006; Padmavati, Thara, & Corin, 2005; Thara, Padmavati, & Srinivasan, 2004).
There is limited literature on EMs of postpartum psychosis when compared to studies on EMs of psychosis in general population. Few studies have focused on barriers to care in postnatal depression, and there is no focus on severe mental illnesses in the postpartum period (Dennis & Chung-Lee, 2006; Goodman, 2009; Ugarriza, 2002). The distress states arising out of common mental disorders during the postnatal period are attributed to vulnerability, supernatural attacks and physical harm (Hanlon, Whitley, Wondimagegn, Alem, & Prince, 2009). Economic difficulties, marital discord, cultural attitudes towards gender have been described as causal explanations of depression in postpartum mothers (Rodrigues, Patel, Jaswal, & de Souza, 2003). Simultaneous use of multiple EMs among rural women with postpartum depression was reported from southern India (Savarimuthu et al., 2010).
The nature of belief systems associated with postpartum psychosis is an under-researched area. Understanding EMs of this condition is particularly important because of the socio-cultural and ritualistic dimensions of the postpartum period in most cultures. The present was undertaken to address the lacunae in understanding EMs of postpartum.
Methodology
The study was conducted at the National Institute of Mental Health and Neurosciences (NIMHANS), Bangalore, a well-known neuropsychiatric care centre in southern India. The hospital provides primary level to tertiary level care for patients with mental illness. Patients were recruited from outpatient and inpatient perinatal psychiatry services by trained research assistants after approval of the study by the Institute Ethics Committee.
Successive patients suffering from postpartum psychosis who agreed to participate in the study were recruited over 2 years after informed consent process. Postpartum psychosis was defined in the study as presence of severe depression, mania or psychosis during the first 6 months following the delivery. International Statistical Classification of Diseases and Related Health Problems–10th Revision (ICD-10) was used for clinical diagnosis, confirmed by two qualified psychiatrists. Patients with mood or psychotic symptoms in association with significant cognitive deficits due to severe brain damage/dysfunction were excluded.
A modified version of the Short Explanatory Model Interview (SEMI) was used to assess patient’s and caregiver’s beliefs about the presenting symptoms and possible causes through open-ended interviews and recorded verbatim. SEMI is short, simple, uses open-ended interviews and can be easily coded for analysis (Bhui & Bhugra, 2002; Lloyd et al., 1998). The names for the illness were elicited through the following questions – What do you call these problems (symptoms)? If you had to give the illness/symptom a name, what would it be?
Causative models of the presenting problems were elicited through the question – Why has it happened (cause)? What do you think has caused this problem?’ Why do you think these problems started when they did (cause)?
Emerging themes were identified from the responses of the patient and accompanying caregivers were coded and entered into Excel software for descriptive analysis. Content analysis of the responses to the questions was done independently by the authors (H.T., A.D.). Differences were resolved through discussions (H.T., A.D., G.D. and P.S.C.), and themes were finalised. Local names for postpartum psychosis were listed and translated to English. These names were then categorised based on common themes. Data were obtained from semi-structured instruments and responses.
Results
A total of 123 women with postpartum psychosis were evaluated along with family members. The mean age of the patients was 25.42 years (standard deviation (SD) = 4.44 years) and the mean number of years of formal education was 9.86 years (SD = 4.22 years). Most patients belonged to either lower (61%) or middle socio-economic status (31.7%), and nearly half (49.6%) of the recruited patients were from a rural background. Most patients were married (99.2%) and majority (77.2%) belonged to Hindu religion. Majority hailed from Bangalore and nearby areas (mean distance of 91.9 km from the psychiatric hospital); 61 (49.6%) patients were primipara. Acute and transient psychotic disorder was the most common clinical diagnosis and was seen in 75 (60.97%) patients. The other diagnoses were depression (14.6%), bipolar disorder (13%), schizophrenia (1.6%) and other psychotic disorders (8.9%). In all, 56 (45.5%) patients had a history of psychiatric illness in the past. Out of 123 interviews during the study, 8 (6.50%) interviews were solely answered by patients, 71 (57.73%) interviews were answered solely by the caregivers, and in 44 (35.77%) interviews, there was an equal participation of both patient and the caregiver.
Names used for postpartum psychosis
In 85 (69%) interviews, the participants had a name for the illness that could be broadly classified into four categories (Table 1) – names specific to mental illness during the postpartum period, names related to mental illness in general, names related to physical causes and names related to possession and supernatural factors constituted the four categories. The word ‘bananti’ was used as a prefix for labelling the distress and refers to a woman in her postpartum period. The term ‘Sanni’ was used to name behavioural changes during the postpartum period. Some names referred to humoral factors such as cold. Few names used to describe postpartum psychosis are also used to name the psychiatric illnesses in general. Names related to possession and supernatural causation translated as ghosts, demons, black magic, ‘bad breeze’ and bad fate. Physical factors as causation were described using terms that referred to worms, general medical problems and nervous system.
Names given for postpartum psychosis by women and caregivers.
EMs based on responses to SEMI
A total of 192 EMs were reported in the present study, with several respondents reporting more than one model (Range = 1–4, Mode = 1, Mean = 1.56, SD = 0.75) (Table 2). Out of 123 interviews, 70 (56.91%) had one EM, 41 (33.34%) had two EMs and 12 (09.75%) had more than two EMs. A biomedical EM of mental illness was seen only in 35.42% of the responses. In all, 23.44% responses suggested psychosocial stressors that included childbirth-related and non-childbirth-related factors as EMs. In all, 20.3% of responses attributed supernatural factors as a cause for the presenting problems. Supernatural factors included black magic, entry of soul and influence of ghosts, evil spirits and evil eye effect; 13.02% of responses did not reveal any particular idea about the cause.
Explanatory models of postpartum illness.
The following narrative accounts illustrate the different types of EMs.
Psychosocial causes – interpersonal problems
Patient Ms SK (26 Years, Urban, Acute and Transient Psychotic Disorder): I continued to stay at my maternal place even after marriage. My husband works as a rickshaw driver and has very low income. I happen to be the second wife to my husband. He is not legally divorced from first wife. After the delivery of baby, my husband called his first wife to take of care me and baby. She would fight with me saying that she is the legitimate wife and I don’t have any right to stay with him. I used to cry all the time.
Patient Ms M (25 years, urban area, Acute and Transient Psychotic Disorder): My husband would beat me every day. He would pull my hair, kick and hit on head. I am now having these problems because of those incidents.
Mother of Ms M: After her delivery both husband and wife fought with each other and he hit her very badly and left her at my place. Following this incident she started behaving abnormally.
Psychosocial causes – poor support
Patient Ms SB (25 years, urban area, Acute and Transient Psychotic Disorder): I have to take care of both baby and family. No one is there to help me as ours is nuclear family. I have to do house hold work, look after the baby and have to work at a garment factory. I feel very tired, don’t feel like interacting with others.
Psychosocial causes – gender of the baby
Patient Ms D1 (25 years Urban, Acute and Transient Psychotic Disorder): I cried after realizing that I have given birth to female baby. I wanted a male child. My family also wanted a male child. Few weeks following the delivery, my sister in-laws made fun by saying that the baby looks very thin and weak. I would then as to think whether I have not fed my baby properly?
Relative of Patient Ms D2 (35 years, Rural, Acute and Transient Psychotic Disorder): This delivery is after 8 years after the 3rd child was born. They have vast agricultural lands and property but there is no one to take care of it in future. They now have four female children and they desperately wanted a male child.
Psychosocial cause – death of first child
Patient Ms RB (28 years, Rural, Severe Depression): I have lost my 1st child. I feel that even to this child something bad may happen and die. I am very worried and sad.
Supernatural causes. Patient Ms S (24 years, Rural, Depression):
Somebody has done maata mantra (black magic) because of which all these problems have started.
Mother of patient Ms M (30 years, Urban, Acute and Transient Psychotic Disorder): I think a ‘devva’ (evil spirit) of a dead person has got into her body. We did some pooja, yantra, mantra etc. (religious rituals) to drive away the evil spirit. The faith healer hit her with the leaves of a neem tree. She said ‘yes’ when he asked her about the soul of some dead person had entered her body. We also took her to a ‘Dargah’ (place of worship). She was given holy ash but she did not improve.
Physical causes – cold
Mother of Patient Ms N (20 years, Rural, Acute and Transient Psychotic Disorder): This problem has occurred because of the effect of sheetha (cold).
Personality factor with psychosocial stressor
Mother of Patient Ms T (21 years, Rural, Acute and Transient Psychotic Disorder): By nature she is a very sensitive person, she does not mingle with anyone and does not speak to anyone unless there is a compelling reason. Her husband troubles her too much. He beats her very badly and confines her to the house. She was all right for about one and a half month following delivery when she began to behave strangely. She could not manage because of her sensitive nature.
Spiritual and religious causes
Mother of Ms S (32 years, Rural, Acute and Transient Psychotic Disorder): I had made a vow to the God but was unable to fulfil the vow. Probably that is the reason for my daughter getting this problem.
Discussion
EMs influence the naming and help seeking for the problems associated with psychiatric illness. It was interesting to note in our study that postpartum psychosis was described by more than 30 different names. While certain names were related to mental illness in general, several names used to label the problems such as ‘bananti sanni’ clearly referred to the postpartum period. Most names used by the subjects could be linked to the underlying EMs, for example, use of term Gaali (Bad breeze) suggested a supernatural cause.
Our study assessed EMs of postpartum psychosis among patients and caregivers and found that mental illness model accounted for only one-third of the EMs despite 50% of the study subjects having a previous history of psychiatric illness. This may be due to poor awareness about mental illness in the community. Psychosocial stressors specifically related to the postpartum period (female baby, premarital pregnancy and death of an infant) and stress due to strained interpersonal relationship with spouse and in-laws emerged as the second most common reported model. Preference for a male baby is a well-known phenomenon in India, and women face an enormous pressure to deliver a male baby (Pande & Malhotra, 2006). Interpersonal problems emerged as a prominent theme within the category of psychosocial factors. Psychosocial issues such as interpersonal problems have been frequently reported as EMs in schizophrenia from India (Srinivasan & Thara, 2001). Beliefs in supernatural factors as the cause of mental illness have been reported in many cultures (Pfeifer, 1994). Some of the supernatural factors such as black magic have been reported as causative factors for schizophrenia in earlier studies from India. Our findings are similar to an earlier study on beliefs about causation in schizophrenia that showed low reporting (12%) of supernatural causes as aetiological factors and are in contrast to a study that reported supernatural causes (70%) as the predominant model in patients with Schizophrenia (Saravanan et al., 2007; Srinivasan & Thara, 2001).
Among the physical factors, effect of cold (sheetha) was the most common cited cause for postpartum psychosis. It is a common practice by families in India to ensure that the postpartum woman is covered in warm clothes to prevent cold exposure. Behaviour changes during postpartum are likely to be attributed to impaired temperature homeostasis if a woman has had exposure to cold.
Personality factors such as excessive worrying, sensitive, anxious, poor adaptability, quarrelsome and stubbornness were attributed as causative factors in few patients. Manifestations of postpartum psychosis are polymorphic and are characterised by irritability and mixed affective symptoms (Ganjekar, Desai, & Chandra, 2013). Hence, it is likely that some of the symptoms may appear to family as exaggeration of personality traits in the patient.
Spiritual factors have been frequently cited as causes for mental disorders in primary care setting (Patel, Gwanzura, Simunyu, Lloyd, & Mann, 1995). However, in our study, spiritual and religious factors contributed very little for EMs.
It is interesting to note that there were a total of 192 EMs reported in our study implying that many participants had more than one EM. Most patients with multiple EMs held two models and some subjects held up to four models. Earlier studies have also reported of multiple EMs for psychosis (Bhikha, Farooq, Chaudhry, & Husain, 2012). Our study found that patients could hold models that were diametrically opposite – such as mental illness and supernatural factors. This may be possible because EMs are shown to be fluid and changeable in nature depending on individual and their socio-cultural circumstances (Kleinman, 1980; McCabe & Priebe, 2004).
EMs are associated with help-seeking behaviour, treatment acceptance and compliance to treatment and have important implications on patient satisfaction with the treatment and on therapeutic relationships (Bhui & Bhugra, 2002; Foulks, Persons, & Merkel, 1986; McCabe & Priebe, 2004). A decision to seek a particular type of care is a complex interplay of individual/household behaviour, community norms, EMs, awareness, expectations, response to treatment, economic and socio-cultural factors (Olenja, 2003). Hence, instead of a single kind of care, it is likely that patient and family would seek care from different types of health-providers. It would be important to study the relationships between EMs and treatment seeking in women with postpartum psychosis. There is a need for further research in this area across diverse socio-cultural backgrounds. The relationship between EMs and duration of untreated illness, the stability of EMs after treatment, the belief systems about postpartum psychosis among traditional faith healers and the impact of postpartum mental health literacy among mothers and caregivers are some of the areas that need to be addressed in future studies.
Strengths and limitations
There was detailed and systematic assessment of EMs associated with postpartum psychosis using a structured instrument. Inclusion of both patient–caregiver as participants in the study and thematic analysis by multiple raters adds credence to the findings. The study sample probably had a severe postpartum psychotic illness that needed treatment from a specialist service. The high rates of past psychiatric disorder may have also influenced the results.
Conclusion
Our study has demonstrated that patients with postpartum psychosis and their caregivers hold multiple EMs of illness that are associated with unique socio-cultural factors. In view of the potential seriousness of the postpartum psychosis, the relationship between EM and treatment-seeking behaviour needs further research. It would be further interesting to study the stability of EMs after treatment in women with postpartum psychosis. An understanding of EMs of postpartum psychosis would help to formulate health education and service delivery for treatment of this condition.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research reported in this publication was supported by the Indian Council of Medical Research, New Delhi, India, under grant number 21/12/09/09/HSR.