Anxiety and Depression, Not Substance Use, Tracked With Symptom Severity in a New Early Schizophrenia Study
| Audience | Psychiatric and primary care clinicians managing patients with schizophrenia-spectrum illness, cannabis clinicians whose patients carry a co-occurring psychotic disorder, and family members of patients with early-phase schizophrenia |
| Primary Topic | A cross-sectional observational study of 170 patients with schizophrenia of less than 5 years’ duration in Northern India, examining how commonly anxiety, depressive symptoms, and substance use (including cannabis) occurred, and whether each correlated with PANSS-rated illness severity, published in Primary Care Companion for CNS Disorders in August 2026 |
| Source | Read the study in Primary Care Companion for CNS Disorders |
Anxiety and Depression, Not Substance Use, Tracked With Symptom Severity in a New Early Schizophrenia Study
A cross-sectional study of 170 patients with early-phase schizophrenia in Northern India found that 32.9% had moderate to severe anxiety and 16.5% had moderate to severe depressive symptoms, and that both correlated significantly with PANSS illness-severity scores. Substance use was common, tobacco (31.2%), alcohol (12.9%), cannabis (11.8%), and opioids (5.3%), but did not correlate highly with illness severity as a group. Full text of the study was not accessible for this review; all claims here are confined to what the published abstract states.
| Study Type | Cross-sectional observational study |
| Setting | Department of psychiatry, a tertiary care teaching hospital in Northern India |
| Study Period | November 2023 to May 2025 (18 months) |
| Population | 170 patients aged 18 to 60 years with schizophrenia of less than 5 years’ duration, diagnosed per ICD-11 criteria |
| Assessment Tools | Positive and Negative Syndrome Scale (PANSS); Hamilton Anxiety Rating Scale (HAM-A); Hamilton Depression Rating Scale (HAM-D); WHO Alcohol, Smoking and Substance Involvement Screening Test, Version 3.0 (WHO ASSIST 3.0) |
| Moderate-to-Severe Anxiety | 32.9% of patients (HAM-A score of 25 or higher) |
| Moderate-to-Severe Depression | 16.5% of patients (HAM-D score of 20 or higher) |
| Substance Use Prevalence | Tobacco 31.2%, alcohol 12.9%, cannabis 11.8%, opioids 5.3% |
| Substance Use and Severity | Substance use did not correlate highly with schizophrenia severity as measured by the PANSS |
| Anxiety and Severity | Significant positive correlations between HAM-A scores and PANSS total, negative, and general psychopathology scores |
| Depression and Severity | Depressive symptoms (HAM-D) correlated moderately with all PANSS components (rho = 0.242 to 0.439, P < .001) |
| Trial Registration | Clinical Trials Registry-India, CTRI/2023/11/059495 |
| Journal | Primary Care Companion for CNS Disorders, 2026;28(4):26m04210 |
| DOI | 10.4088/PCC.26m04210 |
| PMID | 42628038 |
| Authors | Prabhas Pandey, Ajeet Sidana, Prinka Arora |
| Sourcing Note | This review is based on the published PubMed abstract; full text was not accessible through PubMed Central or an open-access repository at the time of writing |
The study team conducted an 18-month cross-sectional observational study, from November 2023 to May 2025, in the psychiatry department of a tertiary care teaching hospital in Northern India, enrolling 170 patients aged 18 to 60 with schizophrenia of less than 5 years’ duration diagnosed under ICD-11 criteria.
Every patient was assessed with the Positive and Negative Syndrome Scale (PANSS) for overall illness severity, the Hamilton Anxiety Rating Scale (HAM-A) and Hamilton Depression Rating Scale (HAM-D) for comorbid mood and anxiety symptoms, and the WHO Alcohol, Smoking and Substance Involvement Screening Test, Version 3.0 (WHO ASSIST 3.0) for substance use. The study was formally registered with the Clinical Trials Registry-India (CTRI/2023/11/059495).
Moderate to severe anxiety, defined as a HAM-A score of 25 or higher, was present in 32.9% of patients, and moderate to severe depressive symptoms, a HAM-D score of 20 or higher, were present in 16.5%.
Both symptom clusters correlated significantly with how severe a patient’s illness looked on the PANSS. Anxiety scores showed significant positive correlations with PANSS total, negative, and general psychopathology scores, while depressive symptom scores correlated moderately with all PANSS components (rho = 0.242 to 0.439, P < .001).
Substance use was frequent in this cohort: tobacco use in 31.2% of patients, alcohol in 12.9%, cannabis in 11.8%, and opioids in 5.3%, as screened with the WHO ASSIST 3.0.
Despite how common substance use was, the abstract reports that it did not correlate highly with schizophrenia severity as measured by the PANSS. The published abstract presents this as a finding for substance use as a pooled group rather than reporting a separate statistical test for cannabis alone, so this study does not establish that cannabis specifically has no relationship to symptom severity, only that substance use overall did not track closely with severity in this sample.
This is a cross-sectional study, capturing a single point in time for each patient, so it cannot establish whether substance use, anxiety, or depression came before or after increases in illness severity, or rule out that all three are being driven by some shared underlying factor.
The sample was drawn from one tertiary care teaching hospital in Northern India, which limits how confidently these prevalence figures and correlations generalize to other regions, health systems, or patient populations, including cannabis medicine patients managed outside a hospital psychiatry setting. Full text of the study was not accessible for this review, so this summary is confined strictly to what the published abstract reports and does not include any dose-response, subgroup, or mechanism detail beyond it.
The study authors conclude that their findings highlight the critical need for early screening and comprehensive management of anxiety and depressive symptoms in schizophrenia, noting that the presence of these comorbidities significantly correlates with illness severity and may influence long-term functional outcomes.
Comorbid anxiety, depression, and substance use are well documented across the schizophrenia literature globally, but this study adds a data point from a population, early-phase schizophrenia patients in India, that the authors describe as less studied than schizophrenia comorbidity in general.
For clinicians who split their attention between substance use counseling and mood or anxiety symptom management in patients with psychotic disorders, this study is a reminder that the two are not automatically equivalent in their relationship to how severe a patient’s current presentation looks, at least in this sample and using this study’s measurement approach.
I see the appeal of this finding, and I want to be careful not to overstate it. In my own practice, when a schizophrenia-spectrum patient is also using cannabis, the reflexive move is often to treat the substance use as the primary lever to address, sometimes ahead of screening for anxiety and depression. This study is a useful, evidence-based check on that reflex: in this particular cohort, substance use as a group did not track closely with how severe a patient’s illness looked, while anxiety and depression did, and did so with real statistical significance.
What I will not do is extend this into a claim that cannabis use is fine for a patient with schizophrenia, or that substance use can be deprioritized. The abstract does not test cannabis on its own, it is one cross-sectional study from a single hospital, and it cannot tell me whether a specific patient’s cannabis use is helping, hurting, or unrelated to their symptom trajectory over time. What it does support is treating anxiety and depression screening in schizophrenia-spectrum patients as at least as clinically urgent as substance use counseling, not as an afterthought to it.
How to Read a Cross-Sectional Comorbidity Study Without Overreaching
A finding that substance use ‘did not correlate highly’ with illness severity can be misread as proof that substance use does not matter in schizophrenia.
Four checks keep this study’s real, more limited contribution in view.
A Four-Step Reading Frame
Separate 'did not correlate with severity' from 'is harmless'
The study measured a statistical association with PANSS severity scores at a single point in time, not whether substance use affects relapse risk, treatment adherence, functioning, or other outcomes it did not measure.
Notice substance use was analyzed as one pooled group
Tobacco, alcohol, cannabis, and opioids were assessed together against severity in the abstract; this study does not report a cannabis-specific correlation separately.
Remember cross-sectional means one snapshot in time
The design cannot establish whether anxiety, depression, or substance use came first, or whether all three trace back to a shared underlying cause.
Hold the single-site limitation in view
All 170 patients were drawn from one tertiary care teaching hospital in Northern India, which limits how confidently the findings generalize elsewhere.
What This Study Means Depending on Who Is Asking
Patients, clinicians, and researchers may read a null finding for substance use very differently. These perspectives examine what this single-site comorbidity study supports for each audience without extending it beyond what the abstract states.
This Study Does Not Give You a Pass on Substance Use
If you live with schizophrenia and also use cannabis or another substance, this study does not mean your substance use is fine or unimportant to your care team; it found that in one hospital-based sample, substance use as a group did not track closely with illness severity, while anxiety and depression did.
If you are experiencing anxiety or low mood alongside a schizophrenia diagnosis, this study is a reminder that those symptoms deserve direct attention from your care team, not just secondary attention behind substance use counseling.
Screen for Anxiety and Depression as Rigorously as Substance Use
For primary care clinicians co-managing patients with schizophrenia, this study supports building routine anxiety and depression screening into visits with the same rigor typically applied to substance use screening, rather than treating mood and anxiety symptoms as secondary.
The correlation coefficients reported here (rho = 0.242 to 0.439 for depression) are moderate, not dramatic, and this remains one cross-sectional study, so it should inform screening priorities rather than replace an individualized clinical assessment.
A Reason for Caution Before Reading Too Much Into This Finding
Cannabis clinicians who see patients with a co-occurring schizophrenia-spectrum diagnosis should be careful not to read this study as reassurance that cannabis use is unrelated to symptom severity; the abstract analyzed substance use as a pooled category, not cannabis specifically, so no cannabis-specific conclusion can be drawn from it.
What is clinically useful here is the reminder that anxiety and depression screening deserves equal weight to substance use discussions in this population, since untreated mood and anxiety symptoms correlated significantly with how severe a patient’s presentation looked.
Comorbidity Burden in Early-Phase Illness Deserves Direct Attention
For psychiatric clinicians, the prevalence figures alone are clinically relevant: roughly a third of patients within 5 years of a schizophrenia diagnosis had moderate to severe anxiety, and about one in six had moderate to severe depression, in a sample where these comorbidities significantly tracked with overall illness severity.
The finding that substance use did not correlate as strongly invites further, more granular research, including studies that separate cannabis from other substances, rather than a conclusion that substance use can be set aside in comorbidity assessment.
A Pooled Substance-Use Variable Limits What Can Be Concluded
Addiction medicine clinicians will likely want more granularity than this abstract provides; tobacco, alcohol, cannabis, and opioids carry different risk profiles and are used at different rates (31.2%, 12.9%, 11.8%, and 5.3%, respectively) in this cohort, and pooling them into one substance-use variable obscures whether any single substance behaves differently from the group average.
This study is a call for future research that tests each substance separately against illness severity, rather than a finding that substance use in general can be treated as clinically neutral in schizophrenia.
Comorbidity Screening Gaps Are the Real Public Health Signal Here
The public health takeaway from this study is less about substance use specifically and more about the scale of undertreated anxiety and depression in early-phase schizophrenia, present in roughly a third and one-sixth of patients respectively in this sample.
Coverage that frames this as ‘cannabis does not worsen schizophrenia’ would overstate what a pooled, cross-sectional, single-site finding can support; the more defensible message is that mood and anxiety comorbidity in schizophrenia needs more systematic screening.
A Null Finding for Substance Use Is Not the Same as Proof of No Relationship
A cross-sectional study finding that substance use ‘did not correlate highly’ with PANSS severity is a null or weak finding, not evidence of absence; with 170 patients and substance use rates as low as 5.3% for opioids, statistical power to detect a real cannabis-specific association may simply have been limited.
The abstract does not report confidence intervals, effect sizes, or the specific statistical test used for the substance-use comparison, which makes it difficult to judge how strong or weak this null finding actually is from the published abstract alone.
A Well-Instrumented Study With Real Design Limits
This study’s strengths are real: a registered protocol, validated instruments (PANSS, HAM-A, HAM-D, WHO ASSIST 3.0), a focus on early-phase illness, and reported correlation statistics rather than only descriptive prevalence.
Its limitations are also real: a single-site, cross-sectional design; a pooled rather than substance-specific analysis of substance use; and, for this review, abstract-only sourcing, since full text was not accessible through PubMed Central or an open-access repository at the time of writing. Any dose-response, subgroup, or mechanistic detail that may exist in the full paper is not reflected here.
Join the Conversation
Have a question about how this applies to your situation? Ask Dr. Caplan
Want to discuss this topic with other patients and caregivers? Join the forum discussion
Frequently Asked Questions
What did this study examine?
A cross-sectional observational study of 170 patients aged 18 to 60 with schizophrenia of less than 5 years' duration, conducted at a tertiary care teaching hospital in Northern India, examined how common anxiety, depressive symptoms, and substance use were, and whether each correlated with illness severity measured by the PANSS.
How common were anxiety and depression in this sample?
Moderate to severe anxiety (HAM-A score of 25 or higher) was present in 32.9% of patients, and moderate to severe depressive symptoms (HAM-D score of 20 or higher) were present in 16.5%.
Did anxiety and depression correlate with illness severity?
Yes. Anxiety scores showed significant positive correlations with PANSS total, negative, and general psychopathology scores, and depression scores correlated moderately with all PANSS components (rho = 0.242 to 0.439, P < .001).
How common was substance use, including cannabis?
Substance use was common: tobacco in 31.2% of patients, alcohol in 12.9%, cannabis in 11.8%, and opioids in 5.3%, screened using the WHO ASSIST 3.0.
Did substance use correlate with illness severity?
No. The study reports that substance use, assessed as a pooled group across tobacco, alcohol, cannabis, and opioids, did not correlate highly with schizophrenia severity as measured by the PANSS.
Does this mean cannabis use does not affect schizophrenia symptoms?
No. The published abstract analyzed substance use as one combined category rather than testing cannabis separately, so this study cannot establish a cannabis-specific conclusion, only that substance use overall did not correlate strongly with severity in this sample.
Can this study show that anxiety or depression cause worse psychotic symptoms?
No. This is a cross-sectional study capturing a single point in time, so it can show correlation but cannot establish which came first or rule out a shared underlying cause.
How generalizable are these findings?
The findings come from one tertiary care teaching hospital in Northern India, so how well the prevalence rates and correlations generalize to other regions, health systems, or patient populations is not established by this single-site study.
Was the full text of this study reviewed for this article?
No. Full text was not accessible through PubMed Central or an open-access repository at the time of writing, so this review is based strictly on the published PubMed abstract, and no dose-response, subgroup, or mechanistic detail beyond the abstract is included.
What do the study authors recommend?
The authors conclude that their findings highlight the critical need for early screening and comprehensive management of anxiety and depressive symptoms in schizophrenia, since these comorbidities significantly correlated with illness severity and may influence long-term functional outcomes.
