Consultation-liaison psychiatric service utilization by suicide attempters.

CONTEXT: There is sparse Indian literature on consultation-liaison psychiatric (CLP) service utilization by suicide attempters who generally present first to the accident and emergency departments and are seen by the psychiatrist only after the initial stabilization. AIMS: The aim of the study is to study the psychiatric referral, review, and psychopharmacological intervention rates among suicide attempters and to study factors associated with psychiatric referral. SETTINGS AND
DESIGN: Retrospective, medicolegal case register-based, explorative analysis of suicide attempters presents to a rural tertiary care hospital in south India. SUBJECTS AND METHODS: Eight hundred and twenty-nine medicolegal case files of individuals whose diagnosis was recorded as either suicidal attempt, deliberate or intentional self-harm, and poisoning or hanging were analyzed for different variables. STATISTICAL ANALYSIS USED: Descriptive frequencies, Pearson's Chi-square and logistic regression analysis to know the association of psychiatric referral with different variables, were performed using the Statistical Package for the Social Sciences, version 16.
RESULTS: A little more than half of the suspected suicide attempters (51.4%) were referred for psychiatric review, and majority of those referred (93.7%) were reviewed by the psychiatrist. Psychiatric referral was significantly associated with a relatively younger age, positive past and family history of suicidal behavior, mode of attempt (pesticide poisoning and attempted hanging were more likely to be referred), and a longer duration of hospitalization.
CONCLUSIONS: The low referral rate but an overwhelming review rate among those referred shows that probably referrer factors are responsible for this and so CLP for suicide attempters needs to be strengthened by sensitizing the referring doctors on the importance of the psychiatric referral and the need to avoid discharge within the first 24 h.

INTRODUCTION

Consultation-liaison psychiatry (CLP) is a subspecialty of psychiatry that has aided the shift of psychiatry from mental asylums and hospitals to general hospital settings. CLP provides care to inpatients under nonpsychiatric care. It involves recognizing and treating mental illness in physical illness contexts, coordinating psychiatric and medical care, and staff education. Despite evidence of benefits of CLP for inpatients with psychiatric comorbidities, referral rates from hospital doctors remain low even in those places where psychiatric services exist. A systemic review grouped barriers to psychiatric referral into systemic factors, referrer factors, and patient factors.[1]

When compared to western figures, several Indian researchers have found a lower referral rate (0.15%–3.6%) but higher psychiatric morbidity in screening studies (31%–34.5%).[23] A study on the psychiatric referrals from intensive care unit to the psychiatry department found a low referral rate of 1.67%, given that prevalence of psychiatric morbidity in general hospitals range from 20% to 60%.[4] Suicidal attempts were seen to contribute to 2.5%–13% of emergency psychiatric referrals,[56] while another study found that one-third of those presenting with psychological problems to the emergency room was brought for medicolegal and social reasons.[7] Suicidal attempt as a reason for psychiatric referral has ranged from 1.35% to 50%.[89101112131415] Two studies have found deliberate self-harm to be the most common reason of referral (50% and 33.9%),[1314] and another has found it to be the second leading cause (20.4%).[12]

There is a higher suicide mortality risk for rural and South Indian population.[16] A questionnaire survey of teaching hospitals in south India found patchy psychiatric services for suicide attempters presenting to the casualty department.[17] It is in this scenario of heightened suicide risk for rural south Indian population and the patchy psychiatric services here that we aimed to study if the mental health needs of those presenting after a suicide attempt were being met. Most of the consultation-liaison studies so far have been either related to reasons and profile of psychiatric referrals or screening for psychiatric morbidity in different populations, and one study has clearly shown that psychiatric intervention in CLP is significantly beneficial.[15]

Objectives

To study the percentage of suspected suicide attempters who received psychiatric referral, review, and psychopharmacological interventions

To study the association of psychiatric referral with sociodemographic variables, mode of attempt, and length of hospitalization

To study common psychiatric diagnosis among those reviewed by the psychiatrist.

SUBJECTS AND METHODS

Eight hundred and twenty-nine medicolegal cases admitted at a multidisciplinary tertiary care hospital in rural South India from January 2013 to December 2015, whose diagnosis was recorded as either suicidal attempt, deliberate or intentional self-harm, and poisoning or hanging were included in the study. Burns and drowning cases (because of the ambiguity of whether these were intentional or accidental), fatal attempts and those cases that were initially recorded as suicidal attempt but were on exploration, found to be due to accidental poisoning were excluded from the study. Data from the case files of the included participants were analyzed for different variables. Being a case register-based study, no contact was established with the participants. Identity of the individuals was protected by deidentifying and codifying the case files. Ethical clearance for this study was obtained from the Institutional Ethics Committee.

The Statistical Package for the Social Sciences (SPSS for Windows, Version 16.0. SPSS Inc, Chicago), was used for analysis. Chi-square test was used to test the association between different categorical variables. Step-wise logistic regression analysis with forward likelihood ratio was performed keeping psychiatric referral as the dependent variable and sociodemographic characteristics, mode of attempt, precipitant, and days of hospitalization as independent variables, to find the predictors of psychiatric referral.

RESULTS

Service utilization

Service utilization refers to how many of the suicide attempters had been referred to and were then reviewed by the psychiatrist and what type of pharmacological or nonpharmacological psychiatric intervention they had received and the number of days they had been hospitalized as a result of their attempt. At the time of the study, the psychiatry department of the hospital where this study was conducted had four psychiatrists and facility for 15 psychiatric inpatients with an average bed occupancy of 75%–90% and a daily outpatient department of 30–40 patients.

Figure 1 depicts how many of the 829 suspected suicide attempters were referred to and then reviewed by the psychiatrists and the nature of intervention that they received. While a little less than half of them were not referred at all, an overwhelming majority of those referred had undergone psychiatric review. A little more than half of those referred were started on psychopharmacological agents that included benzodiazepines, antidepressants, antipsychotics, and mood stabilizers. The others received counseling and supportive psychotherapies.

Figure 1

Psychiatric service utilization by suicide attempters

Mean admitted days were 4.73 (standard deviation = 3.819). Two hundred and sixty-three (31.72%) of patients were admitted for 3–7 days and 182 (21.95%) were admitted for <24 h.

Association of sociodemographic characteristics with respect to psychiatric referral

It was found that age, place of residence, and past and family history of suicidal behavior were significantly associated with psychiatric referral. Variables such as gender, marital status, occupation, or socioeconomic status were, however, not found to be significantly associated. Table 1 depicts this information.

Table 1

Sociodemographic variables

Association of psychiatric referral with the mode of suicidal attempt

Those with pesticide poisoning and hanging were significantly more likely and those with overdose of medicines least likely to be referred. Table 2 depicts this information.

Table 2

Association with mode of attempt

Association of psychiatric referral with the length of hospitalization

Table 3 depicts that likelihood of psychiatric referral progressively increased as the number of hospitalized days increased and was least for those who were discharged within 24 h.

Table 3

Length of hospitalization

Variables associated with psychiatric referral

Step-wise logistic regression analysis was performed keeping psychiatric referral as the dependent variable and sociodemographic characteristics, mode of attempt, and days of hospitalization as independent variables. The model obtained was statistically significant (χ2 value = 254.43, df = 17, P < 0.001) and explained 35% of the variance in the dependent variable – psychiatric referral (Nagelkerke R2 = 0.35). Logistic regression indicated that age group, place of residence, occupational status, past and family history of suicidal behavior, and length of hospitalization could predict the psychiatric referral. Table 4 depicts information of the significant predictors.

Table 4

Results of step-wise logistic regression

Psychiatric diagnosis

Of the 399 individuals who were reviewed by the psychiatrist, the attempt was impulsive and no psychiatric diagnosis could be established in 149 (37.2%) of them. The only recorded diagnosis in such cases was intentional self-harm, deliberate self-harm, parasuicide, or impulsive attempt. Table 5 shows the details of the psychiatric diagnosis.

Table 5

Psychiatric diagnosis

DISCUSSION

When compared to studies that found a referral rate of one-third to less than half,[181920] our study shows a slightly better referral rate of 51.4%. A study conducted in eight low- and middle-income culturally diverse countries, including a center in Chennai (India) and Yuncheng (China) found practically no referral to any professional service in these two centers.[18] In contrast to the above studies and our own study, a study from Newcastle, Australia, found that an extremely high percentage (90.9%) of individuals with deliberate self-poisoning received a psychiatric assessment.[21]

In our study, we found a significantly high review rate of 93.7% among the referred population, which mirrored the Fleischmann et al. study's findings that the majority accept consultation once it is offered to them.[18] This shows that even in low- and middle-income countries where one expects stigma among patients to be a major barrier for psychiatric evaluation, systemic and referrer and not patient factors might be contributing to the low referral rates.

Another study found that 79% of the inpatients admitted for self-poisoning were seen by the psychiatrist, compared to only 34% of the casualty attenders showing that hospitalization improved chances of being seen by the psychiatrist.[22] This finding was similar to our study which found that referral rates increased with length of hospitalization and were least for those who were discharged within 24 h. This study also found that overall assessment by the casualty staff was poorer for those who were not referred to the psychiatrist which indicates possible referrer bias and unfavorable attitudes toward those presenting with suicidal attempts which has been the finding in yet another study.[23]

That history of suicidal behavior increases likelihood of psychosocial assessment was found in our and other studies from the United Kingdom and Australia.[19202425]

While we found that those in the second decade of their life had the highest chances and that older age, especially those above 50 years, had lesser likelihood of referral, other studies have found that older age was associated with an increased rate of psychiatric referral.[1926] Our study found no significant association of psychiatric referral with gender, whereas an Australian study found that males had greater odds of being referred,[25] and contrarily another study found that those not referred for aftercare or hospitalization were more likely to be male.[19] Although rural agrarian societies are known for gender discrimination and a patriarchal mindset, it was interesting to note in our study that neither gender nor marital status was associated with an increased likelihood of referral.

Studies have generally found that those associated with more violent methods of self-harm are more likely to be referred for assessment and aftercare.[1925] While none of our participants had used lethal weapons, we found that the relatively more lethal methods of pesticide poisoning and hanging were more likely to be referred than overdose of medicines or poisoning with household agents.

Among Indian studies, those presenting with self-harm but not having any diagnosable psychiatric disorder have shown varying ranges from 3.5%[27] to 95%.[28] Our rate of 37.2% was similar to some other studies which have had 30%–50% in whom no diagnosis could be established.[2930313233] Most Indian studies have found depressive disorders, commonly ranging between 20% and 40% to be the most common diagnosis followed by adjustment disorder.[28293132343536] A few studies have found adjustment disorder to be the most common disorder.[37383940] Two studies have found personality disorders, mainly emotionally unstable personality disorder to be the most common psychiatric diagnosis.[2741] Two studies found the most common psychiatric diagnosis to be alcohol dependence syndrome.[3442] In our study, the leading diagnoses were adjustment disorder and depressive disorders which accounted for almost 20% each with alcohol abuse/dependence and borderline personality disorder being the third and fourth common diagnosis, respectively.

Limitations

Individuals with less lethal modes of attempts such as wrist slashing might not have been included in the register for medicolegal cases, thereby contributing to selection bias. As only limited information was available in this retrospective, case register-based study, the presence of preexisting medical or psychiatric illnesses and their effect on referral could not be assessed. Referral to the psychiatrist and acceptance of referral by the patient are not the same, and as rejection of referral was not specifically recorded, we cannot comment if verbal rejection influenced referral not being documented in the case notes. Finally, findings of our study that was conducted in a single rural tertiary care hospital cannot be generalized to other hospitals.

CONCLUSIONS

That only around half (51.4%) of the suicide attempters were referred for psychiatric referral, but among those referred, an overwhelming majority (93.7%) were reviewed by the psychiatrist indicates that possibly systemic and referrer factors and not patient factors might be responsible for almost half of them not undergoing a psychiatric evaluation. The hybrid model (stationing a psychiatrist in the medical emergency)[43] would be ideal but is currently not feasible in most hospitals. This study implies that CLP has to be strengthened by sensitizing and educating the referring doctors and the emergency teams that a discharge within 24 h is to be avoided and that psychiatric referral is to be ensured for all cases presenting after a suicidal attempt.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.