‘You don’t bring me flowers any more’: an investigation into the experience of stigma by psychiatric in-patients

AIMS AND METHOD A structured interview-based questionnaire was used to measure the number 
of cards and gifts received by 40 people undergoing psychiatric in-patient 
treatment, compared with an age- and gender-matched group of medical 
in-patients. The study also assessed the amount of disclosure of admission and 
diagnosis to family and friends in the two groups. RESULTS The psychiatric patients received about half as many cards as the medical 
patients (60 v . 112). Gifts to the psychiatric patients were often 
practical in nature and seldom included luxury items such as flowers. 
Disclosure of admission for mental illness (compared with the physical illness 
group) was significantly lower, both to family members (139 v . 193, P =0.041) and friends (74 v . 332, P =0.0001). CLINICAL IMPLICATIONS The stigma of mental illness is reflected in the secrecy surrounding 
disclosure of hospital admission and the lack of tokens of support. Clinicians 
should be aware of the resulting sense of isolation and shame, and the 
consequences for mental health in view of reduced social networks increasing 
the risk of future relapse rates. Reduced contact with mentally ill patients 
has implications for society as a whole in maintaining the status quo of 
stigma.

'You don't bring me flowers any more': an investigation into the experience of stigma by psychiatric in-patients

AIMS AND METHOD
A structured interview-based questionnaire was used to measure the number of cards and gifts received by 40 people undergoing psychiatric in-patient treatment, compared with an age-and gender-matched group of medical in-patients. The study also assessed the amount of disclosure of admission and diagnosis to family and friends in the two groups.

RESULTS
The psychiatric patients received about half as many cards as the medical patients (60 v. 112). Gifts to the psychiatric patients were often practical in nature and seldom included luxury items such as flowers. Disclosure of admission for mental illness (compared with the physical illness group) was significantly lower, both to family members (139 v.193, P=0.041) and friends (74 v. 332, P=0.0001).

CLINICAL IMPLICATIONS
The stigma of mental illness is reflected in the secrecy surrounding disclosure of hospital admission and the lack of tokens of support. Clinicians should be aware of the resulting sense of isolation and shame, and the consequences for mental health in view of reduced social networks increasing the risk of future relapse rates. Reduced contact with mentally ill patients has implications for society as a whole in maintaining the status quo of stigma.
Stigma is a negative attribute that marks an individual or group as being unacceptable, unworthy and inferior. Mental illness engenders stigmatising responses in others and leads to discriminatory behaviour, as it carries with it labels of unreliableness, unattractiveness and dangerousness (Pescoscolido et al, 1999;Crisp et al, 2000;Martin et al, 2000;Corrigan et al, 2003). There is evidence of feelings of rejection and attempts at concealment in the relatives of mentally ill in-patients (Phelan et al, 1998;Ostman & Kjellin, 2002), but less is known about the actions of the patients themselves in responding to stigma. Stigma may result in the internalisation of negative beliefs, causing shame and low self-esteem. This process of 'self-stigma'can lead to reluctance to seek help for mental problems and secrecy about the illness (Corrigan & Watson, 2002). We therefore wished to examine the disclosure of hospital admission by psychiatric patients to family and friends, compared with that of medical in-patients.
Hospital admission for physical illness helps to legitimise the sick role, and the sending of 'get well' cards and flowers signals support from the patient's social network. We wished to test the observed perception that psychiatric in-patients receive fewer tokens of support in the form of cards, flowers and other gifts.

Method
A structured interview-based questionnaire study was conducted on two psychiatric and two medical in-patient wards in a large district general hospital in South-west England. Medical patients were chosen from the gastroenterology and respiratory wards, which had a population of patients felt to mirror the profile of admissions to an adult psychiatric unit. In both types of ward, patients are often admitted during an acute deterioration of their illness, and may be admitted on numerous occasions during their lifetime.
A total of 80 patients who were not too acutely ill or disorientated to have the capacity to give written consent were asked to take part. Patients were given an information sheet explaining the aim of the study and that it would take no more than 15 minutes to conduct, with all responses being confidential. Male and female patients were interviewed alternately to ensure even numbers and reduce gender bias. All patients were aged 18-65 years to ensure greater matching between the psychiatric and medical patient groups. No distinction was made between informally or formally admitted patients. Patients admitted to medical wards with the intention of psychiatric review, for instance following deliberate self-harm, were excluded.
Power calculations performed prior to the study indicated that at least 38 participants were needed in each group to detect a threefold difference in the number of cards received by medical patients compared with psychiatric patients. The appropriate hospital trusts and the Plymouth local research ethics committee approved the study.

The questionnaire
The questionnaire recorded age, ethnicity and number of days of hospital stay. Patients were asked about the number of people in their family, excluding themselves but including parents, spouse, children and siblings. Only Bromley & Cunningham Stigma and psychiatric in-patients original papers relatives who were alive at the time of the admission were counted.
Step-parents, stepchildren and stepsiblings were included in the category of 'family'. Less closely related family members and other social contacts were recorded under 'friends'. The patients' understanding of the reason for their admission was recorded in their own words, e.g. 'liver disease', and they were also asked: (a) How many family/friends had they told of their admission? (b) How many family/friends had they told of their diagnosis? (c) If they had not told any or some family/friends could they say why this was?
Patients were asked to count how many cards and gifts they had received from family and friends, and the type of gift received was also recorded.

Statistical analysis
The results were analysed using the Statistical Package for the Social Sciences, version 9. Although the two study groups were similar in age and gender distribution they were not paired, and the Mann-Whitney U test for non-parametric data was used for comparisons between them. A P value of 0.05 was considered significant.

Results
The two patient groups did not differ significantly on any demographic measurement. The psychiatric patient group comprised 21 women and 19 men, compared with equal gender numbers in the medical group (20 of each), and the ages were distributed similarly in both groups: mean age 42.4 years in the medical group and 38.6 years in the psychiatric group (P=0.41). Family size was also similar in both groups: mean 5.7 in the medical group and 5.4 in the psychiatric group (P=0.29), with 4 being the most common number of family members. Ethnicity was almost exclusively White in both groups (39 v. 40), which reflects the area's demographic character. The medical patients reported a range of gastric, hepatic and respiratory conditions and many had a clear idea of the nature of their disease (e.g ulcerative colitis). Within the medical group, 8 patients said they were admitted 'for investigation' after becoming unwell and did not yet know their diagnosis. The psychiatric patients most often reported that they were admitted for depression (13 out of 40). Five patients reported a diagnosis of schizophrenia, but a further 8 had experienced an episode of psychosis. These 8 patients emphatically described themselves as not having schizophrenia, but rather suffering from the effects of illegal drug use. Five patients were unsure of their diagnosis and said they had not been told why they were in hospital. Table 1 shows the number of cards and gifts received by the patients in total and within the subgroups of family and friends. Psychiatric patients received fewer of these 'get well' symbolic tokens, with just over half as many cards in total (60 v. 112); the difference was greatest for cards from friends (25 v. 57, P=0.09), and the probability shows a trend towards significance. Interestingly, the quantity of gifts received is more equal in the two groups (124 v. 164), but the quality differed. Medical patients received flowers, balloons, magazines, trivia books and luxury foods, such as chocolate. Psychiatric patients generally received more practical gifts of toiletries, foodstuffs and tobacco, with some receiving multiple gifts of the same item during their longer stay, e.g. one patient received 20 packets of cigarettes. Only one patient admitted to a psychiatric ward received flowers in this study, compared with 12 medical patients.  Table 2 shows the number of people the participants informed about their admission to hospital and their diagnosis. Despite the lengthier stays of the psychiatric patients, 6 told no one of their admission at all, and this group was significantly less likely to tell family members compared with the medical patients (139 v. 193, P=0.041). They were even less likely to tell extended family members or friends (74 v. 332, P50.0001). Disclosure of diagnosis was much reduced in the psychiatric patient group to both family (63% v. 95% for medical patients, w 2 =10.76, P=0.001) and friends (55% v. 90% for medical patients, w 2 =10.60, P50.001). The psychiatric patients gave varying reasons for non-disclosure, which predominantly related to their experience of (or fear of) stigma by others. Some examples are given in Box 1. Almost all patients reported that cards and gifts were welcome and that they felt they represented positive support. One psychiatric patient blamed his lack of cards on 'people not knowing what to put on a card if you're mad'.

Discussion
There is a real difference in the amount of cards and gifts received by psychiatric in-patients compared with medical in-patients with similar profiles of illness. The trend did not reach significance and this is likely to be due to differences in the length of stay. The psychiatric patients had a much longer time in which to receive cards and gifts (35 days v. 7 days average stay). Similarly, the short admission times of some medical patients (16 patients had stays of 3 days or less) might have limited the amount of cards and gifts they received from those who knew of their admission. Had the study controlled for length of admission, it is likely that these factors would have resulted in a significant difference in the amount of cards and gifts received. This is supported by the calculated 'rate' of cards received, which is almost ten times greater for medical patients than for psychiatric patients (0.42 per day v. 0.05 per day).
This difference in the amount of tokens of support may also be related to the disclosure of admission by the psychiatric patients, which was significantly less than that of medical in-patients of similar age and gender. Although reduced social contact due to the effects of stigma may mean that there are fewer people to disclose admission to, this is unlikely to account for all of the differences found. The study took place on an acute general psychiatric ward with few long-term rehabilitation patients. Psychiatric patients did not report much smaller family sizes, and yet disclosed their illness to fewer family members. Additionally, they often gave clear reasons for deliberately keeping their admission a secret. The responses to qualitative questions led to the conclusion that secrecy is a decision made secondary to the expected outcomes of the stigma of mental illness, rather than solely the result of fewer social contacts.

Clinical implications
Patients are acutely aware of the beliefs of others about mental illness and are afraid of the rejection that the label carries. This study examines part of the experience of being a psychiatric in-patient, and reveals that secrecy and shame play a large part in patients' feelings, influencing their actions. They may be forgoing the support of  . 'People will always be watching me afterwards.' . 'People are afraid.They judge you differently when they know.' . 'I'm afraid I'll be sacked.' family and friends for the presumed protection from stigmatising consequences that disclosure would bring about. Research into methods of reducing stigma points to contact with people with mental illness as being the most reliable method of producing longstanding change in attitudes . It is evident that the individuals in this study did not expect much benefit from this, and it might be that cultural attitudes have to shift much further before such patients feel the risk to be worth it. The involvement of stakeholders in education programmes and service development, particularly of those prepared to tell their story, is a way of creating opportunities for increased contact and greater understanding.
This study provides evidence for what is often thought to be true, namely that at times of crisis psychiatric in-patients receive fewer gestures of support from family and friends. It is of interest that they usually receive gifts of a practical nature, which may be linked to doubts about the validity of the sick role in mental illness (Kendall, 2001).
The challenge of reducing the stigma of mental illness starts with awareness of its existence and consequences for patients and this may require active questioning. Clinicians should be aware of the risk of isolation, reduction in support from a patient's social network and elements of 'self-stigma' that prevent honesty about a patient's situation. Issues of recovery, compliance and risk of relapse are likely to be affected by patients' experiences of stigma.

Declaration of interest
None.