This cross-sectional study was conducted within a period of one year from January 2018 to December 2018, was approved by institutional ethics committee of Datta Meghe Insititute of Medical Sciences, Wardha, Maharashtra, India. (IEC Approval no: DMIMS[DU]/IEC/2017-18/6706 dated 03/10/2017).

The sample size of 100 patients was considered who gave written informed consent for participating in the study. As per Epi-info online calculator for such study sample size with 95% confidence was 91. Allowing for refusal to provide informed consent and other exclusion criteria, sample size of 100 was decided. Convenience sampling method was used.

Patients who were diagnosed as having schizophrenia as per ICD 10 criteria and were above 18 years of age who gave valid informed consent for participation in the study were included. Those who did not give written informed consent and who did not fulfill ICD-10 diagnostic criteria for schizophrenia and those with an organic mental disorder or mental retardation were excluded from the study.

The tools used in the study were:

SPSS version 22.0 was used for statistical analysis. Schizophrenia patients with substance use comorbidity and those without such comorbidity were compared on demographic and disease related characteristics. Chi-square test was used to analyse categorical data and t-test was used to analyse quantitative data. The p<0.05 was considered statistically significant difference.

1. Sociodemographic characteristics of patients:

A sample size of 100 patients was taken for the study. Sociodemographic details are shown in [Table/Fig-1].

2. Disease related characteristics

Most patients had chronic schizophrenia with duration >2 years (83) and average age at onset was 26.8 years. 55% had more than 2 exacerbations. 16% had family history of psychiatric disorder whereas family history of substance abuse was present in 58%. Family history of suicide was present in 6% patients [Table/Fig-2].

2. Prevalence and pattern of substance use

Out of the total 100 patients of schizophrenia, 48% were found to have substance abuse in their lifetime and 52% had no substance abuse in their lifetime.

Among 48% who had substance abuse 26% had alcohol abuse, 42% had smokeless tobacco and 2% were smokers. Cannabis, opioid and other substance use disorder were conspicuous by their absence and for women patient substance use was limited only to smokeless tobacco.

Of the 48 patients who had substance use disorder 32 patients (66.7%) had onset prior to onset of schizophrenia while 13 patients (27.1%) had onset of substance use after onset of schizophrenia. 3 patients (6.3%) had onset of both disorders simultaneously.

Of the 10 females who had substance use disorder 3 (30%) had onset of substance abuse prior to schizophrenia whereas 7 (70%) had onset of substance use after onset of schizophrenia.

Of the 38 males who had substance use disorder 27 (71.05%) had onset of substance abuse prior to schizophrenia whereas 8 (21.05%) had onset of substance use after onset of schizophrenia.

3. Demographic correlates of substance use

Substance abuse in schizophrenia was associated with male gender, older current age and lower socioeconomic class [Table/Fig-3].

4. Disease related parameters and substance use

Patients with substance use had higher mean PANSS positive score and had higher family history of substance abuse [Table/Fig-4].

Twenty six patients who had alcohol use had AUDIT score more than 8 indicating possibility of alcohol use disorder [Table/Fig-5]. More than 2/3 patients had been advised by someone either to abstain or cut down alcohol use. Five patients (19.2%) had eye opener drink almost daily indicating severe dependence. Four patients (15.4%) had binge drinking episodes at least every month.

5. Correlation of AUDIT score with PANSS scores (N=26)

AUDIT score has low negative correlation with PANSS positive (-.202), PANSS negative (-.103), PANSS general psychopathology (-.133) and total PANSS score (-.187).

Although methodology in previous studies has been variable, we had decided to use several study instruments. However pilot study of first 20 patients suggested none of the patients had comorbid substance abuse except alcohol and smokeless tobacco, hence rather than using a questionnaire like DAST10 or assessing severity of substance use, we chose to rely on interview patients as regards their substance use. Few studies have been tabulated on substance use comorbidity in schizophrenia [Table/Fig-6] [9-15].

Prevalence and pattern of substance abuse: ECA study reported a lifetime prevalence of substance abuse in 47% of schizophrenia population [16]. Dixon L reported 48% [17], Lehman AF et al., [18] reported 54%, Kovasznay B et al., [19] reported 43.8%, Aich T et al., [13] reported 54.3%, comorbid substance abuse in their respective schizophrenia populations.

Types of substances abused: In present study, the prevalence of alcohol abuse was 26%, smokeless tobacco was 42% and smokers were 2% of the study sample. Cannabis was conspicuous by its absence. This may be due to difficulty in growing or procuring cannabis in the dry district of Wardha. Also cannabis use is less in Maharashtra compared to the national average according to the national data on magnitude of substance use in India [20].

In the available literature there is no consistency in pattern of drug abuse in patients of schizophrenia. The number of patients vary in studies from 10% to 70%, depending on the place of study, type of population studied, male to female ratio and how schizophrenia and substance use were defined [6].

Fowler IL et al., found that 48.4% of schizophrenia patients abused alcohol while 36% abused cannabis [21]. Aich T et al., in a sample of 70 patients found 22.8% abusing alcohol, 32.8% nicotine, 32.8% cannabis and 2.8% opoid [13].

In the recent National Mental Health Survey of India [2016], The prevalence of tobacco use disorder (moderate and high dependence) and alcohol use disorder (dependence and harmful use/alcohol abuse) was 20.9% and 4.6%, respectively. The prevalence of alcohol use disorders in males was 9% as against 0.5% in females. Survey also revealed that 0.6% of the 18+ population were recognised with illicit substance use disorders (dependence + abuse) which included cannabis products, opioid drugs, stimulant drugs, inhalant substances and prescription drugs. Among adult males this was 1.1%. There was a wide variation across different states, and similar high rates of consumption of illicit drugs were reported by participants in many states during our focused group discussions. The burden of substance use disorders, contributed by alcohol and tobacco showed prevelance in middle aged (40-59 years) individuals (29%), males (35.67%) and in rural areas (24.12%). However, other substance use disorders (illicit drugs) were more prevalent in urban metro areas [20].

Characteristics of patients: There was not much difference between the two groups i.e., patients with substance abuse and patients without substance abuse on various demographic and clinical parameters. Thus the two groups were comparable. The major exceptions were gender and socioeconomic status. Substance abuse was less common in women and it was limited to smokeless tobacco. This finding is consistent with the cultural norms as substance use especially alcohol and smoking is not socially acceptable in rural India. Similar gender difference was seen in studies by Tekin Uludağ Y and Güleç G [10], Archie S et al., [22], Weilbell M et al., [23], Chaves L and Shirakawa I [24], Zhang XY et al., [25]. Thus in general, males are at greater risk for substance abuse.

Difference in socioeconomic status may be related to majority of study population belonging to rural area. According to national data substance use is more common in rural population. Substance abuse was seen more often among schizophrenia patients who lived in urban areas in the study by Aich T et al., [13] and Yee A et al., [15]. In this study preponderance of substance abuse among the rural patients was noted. This may be related to easy availability of the substances like tobacco, smoking and alcohol even in rural areas.

PANSS SCORE: In present study, patients with substance use had higher PANSS positive score compared to those who did not have substance abuse. Similar findings were observed in studies conducted by Krishnadas R et al., [26], Yee A et al., [15], Zhang XY et al., [25]. Study by Batki SL et al., also suggests association of negative schizophrenia symptoms with alcohol and other substance use [27].

Symptom exacerbation and substance use: Westmeyer J, in their study found that substance use disorders negatively affect the clinical status of patients with schizophrenia and the patients end up having poorer outcome [28]. No such correlation was found in present study.

Family History of substance use in schizophrenia patients: Higher incidence of family history of substance use was obtained in study conducted by Cantor- Graae E et al., [29] which was similar to the findings of present study. Aich T et al., [13] had 17% of substance abuse group with a family history of substance abuse, while it was 37% in the study conducted by Yee A et al., [15] on nicotine abuse in schizophrenia.

AUDIT scores and schizophrenia: Studies using AUDIT to screen alcohol use disorder among schizophrenia patients report alcohol use disorder in schizophrenia patients from 8% to 30% [3,10]. Although none of these studies have analysed patients alcohol use disorders item wise. Hence, we do not have figures for comparison.

AUDIT score had low negative correlation with PANSS score which implies that increasing AUDIT scores were associated with lower PANSS positive scale, negative scale, general psychopathology scale and total PANSS score. The well-known four that are widely available are the Dual Recovery Therapy (DRT) approach, modified Cognitive Behavioral Therapy (CBT), modified MET, and the Substance Abuse Management Module (SAMM) [30].

As the sample size of the study was small, its results cannot be generalised. No biochemical test was done to confirm substance use. As this is a cross-sectional study, effect of substance use on evolution of schizophrenia could not be studied.

Although the correlation of AUDIT score and severity of psychopathology of schizophrenia was not statistically significant, this lends support that alcohol use in these patients may be related to an attempt to reduce psychopathology. Several clinical therapy manuals have been established and verified for value of treating schizophrenia and addiction. There is a clear need to implement these therapies to improve the treatment outcome of schizophrenia patients with comorbid substance use.