Evidence Review: Emotional blunting and subjective personality change during SSRI and SNRI treatment in adults with depression or anxiety

Reviewed by Dr. Margarita Krasnova, MD · Last reviewed 2026-05-28

Evidence Review: Emotional Blunting and Subjective Personality Change During SSRI and SNRI Treatment in Adults with Depressive and Anxiety Disorders — A Scoping Review of Available Evidence

This document is presented as a Scoping Review of Available Evidence rather than a fully PRISMA-compliant systematic evidence synthesis. The reviewer assessed the alignment of the retrieved evidence with the prespecified PICO and identified the following limitation: the included studies showed only partial overlap with the target population's prespecified PICO characteristics.

Abstract

Background: SSRIs and SNRIs are first-line treatments for depressive and anxiety disorders, yet patients frequently report emotional blunting, reduced emotional authenticity, or a subjective sense of personality change during treatment. Whether these experiences represent a direct pharmacological effect or residual symptoms of the underlying illness remains poorly delineated, with implications for treatment adherence and shared clinical decision-making.

Objective: This scoping review mapped the available evidence on the association between SSRI or SNRI treatment and emotional blunting, reduced emotional responsiveness, or subjective personality changes in adults with depressive or anxiety-spectrum disorders.

Methods: A systematic search of electronic databases was conducted using PICO-based eligibility criteria targeting adults with depressive or anxiety disorders treated with SSRIs or SNRIs, with outcomes including emotional blunting, affective flattening, quality of life, interpersonal functioning, and treatment adherence. A single reviewer screened, extracted, and synthesized data narratively. Thirty articles met inclusion criteria, spanning randomized trials, systematic reviews, observational studies, case reports, and study protocols.

Results: Only one included study directly measured emotional blunting using a validated instrument (the Oxford Depression Questionnaire), finding that approximately two-thirds of antidepressant-treated adults with self-reported major depressive disorder endorsed blunting symptoms, correlated with worse depression, anxiety, and quality of life; a second study documented high early treatment abandonment rates without isolating emotional blunting as a driver. The remaining 29 articles addressed adjacent questions such as antidepressant efficacy, adverse-event profiling, neuroimaging correlates, or treatment discontinuation without specifically assessing emotional blunting. No controlled trial used emotional blunting as a prespecified primary outcome. High-certainty meta-analytic evidence confirmed antidepressant superiority over placebo for generalized anxiety disorder and OCD, yet none of these syntheses disaggregated emotional side effects. High early treatment discontinuation rates and withdrawal symptoms during dose reduction were documented across several studies, but no study linked discontinuation specifically to emotional blunting.

Conclusions: Available evidence suggests that emotional blunting may be common during SSRI and SNRI treatment, but the near-complete absence of controlled, prospective studies using validated emotional-blunting instruments means that the frequency, severity, and causal attribution of this phenomenon remain uncertain. Purpose-designed trials with emotional blunting as a primary endpoint are needed to distinguish medication-related emotional restriction from residual illness effects.

Keywords

Selective Serotonin Reuptake Inhibitors, Serotonin and Norepinephrine Reuptake Inhibitors, Affective Symptoms, Depressive Disorder, Anxiety Disorders

1. Introduction

Depressive and anxiety disorders — including major depressive disorder, persistent depressive disorder, generalized anxiety disorder, panic disorder, social anxiety disorder, and obsessive-compulsive disorder — rank among the leading causes of disability worldwide and affect a substantial proportion of the adult population. Beyond their well-characterized cognitive and somatic features, these conditions profoundly alter emotional experience: patients frequently describe a narrowing of affective range, diminished capacity for pleasure, or a pervasive sense of emotional disconnection. Because emotional responsiveness is central to interpersonal relationships, occupational functioning, and overall quality of life, any treatment that modifies the emotional landscape of these disorders carries implications that extend well beyond symptom-score reduction.

Selective serotonin reuptake inhibitors and serotonin-norepinephrine reuptake inhibitors — agents such as sertraline, fluoxetine, escitalopram, citalopram, paroxetine, fluvoxamine, venlafaxine, desvenlafaxine, and duloxetine — remain first-line pharmacotherapy for most of these conditions. While their efficacy in reducing core depressive and anxious symptoms is well established, a recurring concern in clinical practice is that serotonergic antidepressants may themselves induce a state of emotional blunting, numbing, or flattened affect that patients experience as qualitatively distinct from their illness. Some individuals report feeling less sad yet simultaneously less capable of joy, empathy, or creative engagement, raising questions about whether the medication has genuinely restored emotional health or merely substituted one form of affective restriction for another. Disentangling medication-attributable emotional dulling from the residual affective symptoms of incompletely treated depression or anxiety has proven difficult, and the existing literature spans heterogeneous study designs, measurement instruments, and comparator conditions. No recent synthesis has systematically appraised this evidence base with explicit attention to the distinction between drug-related and illness-related emotional changes.

This review aims to evaluate whether SSRI or SNRI treatment is associated with emotional blunting, reduced emotional authenticity, or subjective personality changes in adults diagnosed with depressive or anxiety-spectrum disorders, compared with placebo, untreated illness, psychotherapy alone, healthy controls, non-serotonergic antidepressants, different antidepressant doses, or pre-treatment baseline functioning. Outcomes of interest include emotional blunting and numbing, reduced intensity of positive or negative emotions, affective flattening, changes in empathy, motivation, creativity, interpersonal and relationship functioning, quality of life, treatment adherence, and antidepressant discontinuation attributable to emotional side effects.

2. Methods

2.1 Search Strategy

A systematic search of PubMed, OpenAlex, ClinicalTrials.gov, Author's reference collection was conducted using a structured Boolean query assembled from Population, Intervention/Exposure, and Outcome keyword sets derived from the study's PICO framework. The population of interest was: Adults aged 19 years and older diagnosed with major depressive disorder, persistent depressive disorder, generalized anxiety disorder, panic disorder, social anxiety disorder, obsessive-compulsive disorder, or related depressive/anxiety-spectrum disorders. The intervention/exposure of interest was: Treatment with selective serotonin reuptake inhibitors (SSRIs) or serotonin-norepinephrine reuptake inhibitors (SNRIs), including sertraline, fluoxetine, escitalopram, citalopram, paroxetine, fluvoxamine, venlafaxine, desvenlafaxine, duloxetine, or similar serotonergic antidepressants. The primary outcome was: Emotional blunting, emotional numbing, reduced emotional responsiveness, subjective loss of emotional authenticity, affective flattening, reduced intensity of positive or negative emotions, quality of life, interpersonal functioning, relationship satisfaction, creativity, motivation, emotional empathy, treatment adherence, antidepressant discontinuation due to emotional side effects, and depression/anxiety symptom severity. The search was conducted on 2026-05-26. Records were deduplicated before screening (see §2.3 for counts).

Boolean Query (verbatim)

PubMed

text ((("Anxiety Disorders"[MeSH Terms] OR "Generalized Anxiety Disorder"[Title/Abstract]) AND ("adults"[Title/Abstract] OR "young adult"[Title/Abstract] OR "Adult"[MeSH Terms] OR "adult"[Title/Abstract])) AND ("serotonin noradrenaline reuptake inhibitors"[Title/Abstract] OR "Selective Serotonin Reuptake Inhibitors"[MeSH Terms] OR "selective serotonin reuptake inhibitors"[Title/Abstract] OR "Serotonin and Noradrenaline Reuptake Inhibitors"[MeSH Terms] OR "serotonin-norepinephrine reuptake inhibitors"[Title/Abstract] OR "serotonin reuptake inhibitor"[Title/Abstract] OR "selective serotonin reuptake inhibitor"[Title/Abstract] OR "serotonin noradrenaline reuptake inhibitor"[Title/Abstract] OR "serotonin-norepinephrine reuptake inhibitor"[Title/Abstract] OR "serotonergic antidepressants"[Title/Abstract] OR "serotonergic antidepressant"[Title/Abstract] OR "antidepressant medication"[Title/Abstract] OR "antidepressants"[Title/Abstract] OR "antidepressant treatment"[Title/Abstract])) AND 2023:2026[dp]

OpenAlex

text (("Anxiety Disorders" OR "Generalized Anxiety Disorder" OR "anxiety disorder") AND (adults OR "young adult" OR Adult OR adult)) AND ("serotonin noradrenaline reuptake inhibitors" OR "Selective Serotonin Reuptake Inhibitors" OR "selective serotonin reuptake inhibitors" OR "Serotonin and Noradrenaline Reuptake Inhibitors" OR "serotonin-norepinephrine reuptake inhibitors" OR "serotonin reuptake inhibitor" OR "selective serotonin reuptake inhibitor" OR "serotonin noradrenaline reuptake inhibitor" OR "serotonin-norepinephrine reuptake inhibitor" OR "serotonergic antidepressants" OR "serotonergic antidepressant" OR "antidepressant medication" OR antidepressants OR "antidepressant treatment" OR "Serotonin Uptake Inhibitors" OR "antidepressant therapy" OR antidepressant)

ClinicalTrials.gov

text (("Anxiety Disorders" OR "Generalized Anxiety Disorder" OR "anxiety disorder" OR "Major Depressive Disorder" OR "major depressive disorder" OR "persistent depressive disorder" OR "Panic Disorder" OR "panic disorder") AND (adults OR "young adult" OR Adult OR adult)) AND ("serotonin noradrenaline reuptake inhibitors" OR "Selective Serotonin Reuptake Inhibitors" OR "selective serotonin reuptake inhibitors" OR "Serotonin and Noradrenaline Reuptake Inhibitors" OR "serotonin-norepinephrine reuptake inhibitors" OR "serotonin reuptake inhibitor" OR "selective serotonin reuptake inhibitor" OR "serotonin noradrenaline reuptake inhibitor" OR "serotonin-norepinephrine reuptake inhibitor" OR "serotonergic antidepressants" OR "serotonergic antidepressant" OR "antidepressant medication" OR antidepressants OR "antidepressant treatment" OR "Serotonin Uptake Inhibitors" OR "antidepressant therapy" OR antidepressant OR "Antidepressive Agents" OR SSRI)

The Boolean query is assembled as (Population) AND (Intervention/Exposure) only. Comparator and Outcome terms are not AND-joined into the search query, in line with the Cochrane Handbook §4.4.4: outcome and comparator filters belong in inclusion criteria and post-retrieval relevance assessment, not in the search query. This convention prevents premature exclusion of articles that report the outcome or comparator using vocabulary not in the keyword list.

Retrieved records were screened for relevance against the study goal and the prespecified PICO before full-text review and synthesis.

2.1.1 Scope Note (Scoping Review Framing)

This document is presented as a scoping review of available evidence, framed in accordance with the Arksey and O'Malley (2005) scoping review framework and the JBI scoping review methodology. The reviewer assessed the alignment of the retrieved evidence with the prespecified PICO at the end of the retrieval stage and identified the following limitations:

• the included studies showed only partial overlap with the target population's prespecified PICO characteristics

The methodological implication of scoping-review framing is that broader inclusion criteria are accepted: heterogeneous study designs, a wider range of population overlap, and variation in intervention delivery are all tolerated rather than excluded. Pooled effect estimates are not pursued; the synthesis approach is charting and narrative mapping of the available evidence.

2.2 Inclusion and Exclusion Criteria

Population: Adults aged 19 years and older diagnosed with major depressive disorder, persistent depressive disorder, generalized anxiety disorder, panic disorder, social anxiety disorder, obsessive-compulsive disorder, or related depressive/anxiety-spectrum disorders

Intervention/Exposure: Treatment with selective serotonin reuptake inhibitors (SSRIs) or serotonin-norepinephrine reuptake inhibitors (SNRIs), including sertraline, fluoxetine, escitalopram, citalopram, paroxetine, fluvoxamine, venlafaxine, desvenlafaxine, duloxetine, or similar serotonergic antidepressants

Comparator: Placebo, untreated depression or anxiety, psychotherapy-only treatment, healthy controls, non-serotonergic antidepressants, different antidepressant doses, or pre-treatment baseline emotional functioning

Outcome: Emotional blunting, emotional numbing, reduced emotional responsiveness, subjective loss of emotional authenticity, affective flattening, reduced intensity of positive or negative emotions, quality of life, interpersonal functioning, relationship satisfaction, creativity, motivation, emotional empathy, treatment adherence, antidepressant discontinuation due to emotional side effects, and depression/anxiety symptom severity

Articles were included if they met the prespecified PICO criteria and open-access full text was available for synthesis. Full-text availability was assessed against PubMed Central and additional open-access repositories.

Eligibility Constraints

• Search executed 2026-05-26; publication-year filter: 2024–2026.
• No language filter applied at retrieval; all results returned by the database query were considered for screening.
• Study design: no design filter was applied at retrieval; study designs were classified post-retrieval (see §2.3 Study Selection).

2.3 Study Selection

1730 records were identified across the searched databases (PubMed: 176, OpenAlex: 1053, ClinicalTrials.gov: 500, Author's reference collection: 1). After deduplication, 1696 records remained (from 1730 records prior to deduplication). Following title and abstract screening, 820 records were assessed for relevance. After relevance assessment, 820 articles met the inclusion criteria. Of these, 30 had retrievable full text and were included in the narrative synthesis.

2.4 Data Extraction

Data were extracted from each of the 30 included articles by a single reviewer using a structured extraction template. For every study, the reviewer recorded the main finding, the reported direction of effect on emotional blunting or related affective outcomes, study design, population characteristics, follow-up duration, sample size, effect-magnitude language as stated by the original authors, and author-stated limitations. Effect-magnitude language was preserved in the authors' own terms rather than re-coded, so that hedged or qualified claims retained their original strength. Where a study reported on more than one relevant outcome — for instance, both reduced emotional responsiveness and changes in interpersonal functioning — each outcome was captured separately within the same extraction record. No second reviewer independently verified the extractions; this single-reviewer approach is disclosed as a methodological consideration in the limitations discussion.

2.5 Risk of Bias Assessment

Risk of bias was assessed using QUADAS-2 and ROBINS-I and RoB 2.

Article	Instrument	Overall judgment
Sem Cohen 2025 (PMC 12404328)	QUADAS-2	Some concerns
Shalini Shukla 2025	ROBINS-I	High
Andrea Cipriani 2026 (PMC 12961600)	RoB 2	High
Jeffrey A Mills 2024 (PMC 11807899)	ROBINS-I	Some concerns
Tempei Otsubo 2025 (PMC 12683617)	RoB 2	Some concerns
Dheeraj Rai 2024 (PMC 10782796)	RoB 2	Unclear
Laura Orsolini 2024 (PMC 10965133)	ROBINS-I	High
Anouk van der Straten 2024 (PMC 11918735)	ROBINS-I	High
Gizem Nur Pala Avan 2025 (PMC 11680499)	ROBINS-I	High
Yiding Han 2025 (PMC 12304266)	ROBINS-I	Some concerns
Taro Kishi 2025 (PMC 13040590)	ROBINS-I	Some concerns
Sem E Cohen 2025 (PMC 12492065)	ROBINS-I	Some concerns
Prashant Babubhai Patel 2026	ROBINS-I	High
Komal Kohat 2026 (PMC 12998030)	ROBINS-I	High
Liang Xue 2026 (PMC 12929608)	ROBINS-I	Some concerns
Hiroe Hu 2024 (PMC 11581486)	RoB 2	Some concerns
Rachael Frost 2026	ROBINS-I	Unclear
Siegfried Kasper 2024	RoB 2	Some concerns
Andri Rennwald 2026	ROBINS-I	High
Katarina Kopcalic 2025 (PMC 11779548)	ROBINS-I	Some concerns
Judith Cukor 2024	ROBINS-I	High
T. Jiménez Aparicio 2024	ROBINS-I	High
Hasan Mirzazadeh 2025 (PMC 12131613)	RoB 2	Some concerns
Francesco Monaco 2025 (PMC 12821398)	ROBINS-I	High
Jurriaan M J L Brouwer 2024 (PMC 11129450)	ROBINS-I	Some concerns
Toshiaki Kikuchi 2024 (PMC 11144621)	ROBINS-I	High
Sucharita Mandal 2025 (PMC 12642335)	RoB 2	Unclear
Amanda Hazel Dilmore 2025 (PMC 12092254)	ROBINS-I	High
Luca Giacovelli 2025 (PMC 12459139)	ROBINS-I	High
Diana Dubrall 2025	ROBINS-I	High

2.6 Synthesis Methods

This document is presented as a scoping review of available evidence, synthesized following the Arksey and O’Malley (2005) scoping review framework and the JBI scoping review methodology. The synthesis approach is charting and narrative mapping rather than meta-analysis or pooled effect estimation. Studies were charted according to direction of reported effect, study design, and population characteristics; findings are reported descriptively across the structured narrative subsections of Section 3.2. The scoping framework was selected because the reviewer judged the retrieved studies to have insufficient overlap with the prespecified PICO to support a standard systematic review synthesis. Broader inclusion criteria are therefore accepted, accommodating heterogeneous study designs and a wider range of population overlap than a tightly-specified systematic review would require.

PRISMA 2020 Flow Diagram

The flow of records through identification, screening, eligibility, and inclusion is summarized below (PRISMA 2020 Item 16a).

Identification

1730 records were identified.

• PubMed: 176
• OpenAlex: 1053
• ClinicalTrials.gov: 500
• Author's reference collection: 1

Screening

Records removed as duplicates: 34

Records after deduplication: 1696

Records screened: 1696

Records excluded during screening: 876

Eligibility

Records assessed for eligibility: 820

Records excluded at eligibility: 0

Included

Records excluded (no retrievable full text): 790

Studies included in synthesis: 30

3. Results

Coverage and Validity Scorecard

Field	Value
Databases searched	PubMed, OpenAlex, ClinicalTrials.gov, Author's reference collection
Articles identified per database	PubMed: 176, OpenAlex: 1053, ClinicalTrials.gov: 500, Author's reference collection: 1
Total identified (pre-dedup)	1730
Articles after deduplication	1696
Records after title and abstract screening	820
Records after eligibility assessment	820
Relevance filter	None applied
Articles in synthesis	30
PRISMA 2020 checklist coverage	25 ADDRESSED · 2 PARTIALLY (Items 7, 14 — single-reviewer screening; no PROSPERO pre-registration) · 0 NOT ADDRESSED
Risk-of-bias distribution	Some concerns: 12, High: 15, Unclear: 3
GRADE final certainty	Very Low

All values above are counts reflecting the records retrieved and screened for this review. No composite quality or validity scores are computed; reviewers are expected to interpret these counts in context.

3.1 Study Characteristics

Study	Design	Sample Size	Population	Intervention	Comparator	Follow-up	Primary Outcome	Direction of Effect	Effect Estimate
Sem Cohen, 2025	Diagnostic test accuracy meta-analysis with individual participant data	1753	Adults diagnosed with OCD using DSM-III, III-R, or IV criteria, randomized to active treatment (SSRIs or clomipramine) in double-blind, placebo-controlled, industry-sponsored short-term trials submitted to the Dutch Medicines Evaluation Board; mean age 38.2 years, 54% male, mean baseline YBOCS 24.5	SSRIs or clomipramine treatment	Placebo	10–13 weeks	Early non-improvement predicted subsequent nonresponse in OCD patients with PPV of 86%	Mixed or inconclusive	PPV: 86 (95% CI 83–88)
Shalini Shukla, 2025	Prospective observational study	45	Drug-naïve adult patients aged 18-60 years with newly diagnosed generalized anxiety disorder (ICD-10 criteria) prescribed SSRI monotherapy, attending the psychiatry outpatient department of a tertiary care hospital in rural Mewat region, Haryana, India	SSRI monotherapy	NR	6 weeks	SSRIs demonstrate acceptable safety profile in GAD with predominantly mild adverse drug reactions	Mixed or inconclusive	NR
Andrea Cipriani, 2026	RCT	493	Adults aged 18 to 74 years with major depressive disorder, recruited from 47 sites in Brazil, Canada, and the UK (median age 35 years; 58% female; mean PHQ-9 score 16.6; mean GAD-7 score 11.5)	PETRUSHKA clinical decision-support tool for personalized antidepressant selection	Usual care	24 weeks	PETRUSHKA tool reduced treatment discontinuation and improved depressive and anxiety symptoms	Mixed or inconclusive	adjusted relative risk: 0.62 (95% CI 0.44–0.88, p=.007)
Jeffrey A Mills, 2024	Bayesian hierarchical meta-analysis	82	Children, adolescents, and adults with anxiety disorders (generalized, separation, and social anxiety disorders as well as panic disorder) enrolled in prospective, randomized, parallel-group, placebo-controlled trials of SSRIs and SNRIs	SSRIs and SNRIs	Placebo	NR	Antidepressant-related improvement occurs early; SNRIs produce greater responses in adults than youth	Mixed or inconclusive	mean difference (standardized change in continuous measures of anxiety): -1.98 (95% CI -10.2–14.2, p=0.750)
Tempei Otsubo, 2025	RCT	357	Japanese outpatients aged ≥18 years with a primary diagnosis of generalized anxiety disorder based on DSM-5 diagnostic criteria, HAM-A total score ≥20, CGI-S score ≥4, and GAD-7 total score ≥10	Venlafaxine ER	Placebo	8 weeks (treatment period), plus 2-week tapering period and 4-week follow-up	Venlafaxine ER showed statistically significant superiority over placebo in reducing HAM-A scores	Mixed or inconclusive	LS mean difference: -1.9 (95% CI -3.4–-0.4, p=0.012)
Dheeraj Rai, 2024	RCT protocol (no results reported)	NR	Adults aged 18 years and older with a specialist diagnosis of autism spectrum disorder and a GAD-7 anxiety score of 10 or greater at screening, including those with co-occurring mild intellectual disability	Sertraline	Placebo	52 weeks post-randomisation	Protocol for RCT of sertraline for anxiety in adults with autism; no results	Mixed or inconclusive	NR
Laura Orsolini, 2024	case report	1	A single 45-year-old Caucasian outpatient woman with severe obsessive-compulsive disorder comorbid with anorexia nervosa, restrictive type	Lurasidone add-on to fluoxetine	NR	6 months	Lurasidone add-on to fluoxetine associated with full remission of OCD symptoms	Mixed or inconclusive	NR
Anouk van der Straten, 2024	Longitudinal nonrandomized controlled trial	54	Adults aged 18-70 with a DSM-IV diagnosis of obsessive-compulsive disorder (OCD) recruited from outpatient clinics, plus 20 matched healthy controls	SSRIs or CBT	Healthy controls	16 weeks	SSRIs and CBT lead to primarily opposite changes in brain function in OCD	Mixed or inconclusive	NR
Gizem Nur Pala Avan, 2025	cross-sectional	546	273 adult outpatients diagnosed with depressive disorder or anxiety disorder using antidepressants for at least 1 month, and 273 healthy controls matched for age and gender	Antidepressant use	Healthy controls (no antidepressant use)	NR	Bruxism detected in 73.3% of antidepressant users vs 28.2% of controls	Mixed or inconclusive	OR: 5.928 (95% CI 3.708–9.478, p=< 0.001)
Yiding Han, 2025	Longitudinal observational (pre-post treatment) with healthy control comparison	102	49 drug-naive OCD patients and 53 healthy controls, with 34 OCD patients completing follow-up after 4-week paroxetine treatment	Paroxetine	Healthy controls	4 weeks (mean 33.59 ± 5.05 days)	Paroxetine enhanced bottom-up effective connectivity from striatal subregions to cortical regions in OCD	Mixed or inconclusive	NR
Taro Kishi, 2025	systematic review and pairwise meta-analysis	NR	Individuals with OCD who were stabilized with antidepressants, drawn from double-blind randomized placebo-controlled trials with an enrichment design (n = 1084; mean age 32.8 years; 53.3% male)	Antidepressant maintenance treatment	Antidepressant discontinuation (placebo)	Mean study duration 26.3 weeks; relapse rates compared at 4, 8, 12, 16, 20, and 24 weeks	Antidepressant maintenance significantly lowered relapse rates compared to discontinuation in OCD patients	Mixed or inconclusive	RR: 0.53 (95% CI 0.42–0.68, p=<0.01)
Sem E Cohen, 2025	Individual patient data meta-analysis (IPDMA)	NR	Adults (18 years and over) with a DSM-III, III-R, or IV classification of OCD, enrolled in industry-sponsored short-term, randomised, placebo-controlled SSRI trials	SSRIs	Placebo	10 to 13 weeks	SSRIs are superior to placebo for treating OCD with a small effect size	Mixed or inconclusive	MD: 2.65 (95% CI 1.85–3.46, p=< 0.0001)
Prashant Babubhai Patel, 2026	Prospective observational cohort study (pragmatic, non-randomized)	107	Newly diagnosed adult cancer patients with histologically confirmed solid malignancy diagnosed with MDD or GAD via SCID-5-CV at a tertiary care hospital in India	Combined SSRI pharmacotherapy and CBT; SSRI monotherapy; CBT alone	Each other (three-arm comparison)	6 weeks	Combined SSRI and CBT showed most significant depression improvement in cancer patients	Mixed or inconclusive	partial η²: 0.57 (p=< 0.001)
Komal Kohat, 2026	case report	1	A single 42-year-old married Indian woman diagnosed with generalized anxiety disorder (baseline HAM-A score of 32)	Sertraline (initial); then pharmacogenetic-guided venlafaxine with adjunctive buspirone	NR	NR	Sertraline induced bruxism which resolved after switching to pharmacogenetic-guided venlafaxine	Mixed or inconclusive	NR
Liang Xue, 2026	Prospective cohort study	120	60 drug-naïve patients with first-episode GAD (ICD-10 criteria) and 60 age- and sex-matched healthy controls, all Han Chinese ethnicity, aged 18–65 years	Escitalopram	Healthy controls	8 weeks	Baseline oxytocin and OXTR rs53576 genotype did not predict escitalopram treatment response in GAD	Mixed or inconclusive	Cohen's d: 0.73 (p=< 0.001)
Hiroe Hu, 2024	Secondary analysis of a randomized clinical trial	276	Community-dwelling adults with a primary anxiety disorder (agoraphobia, panic disorder, GAD, or social anxiety disorder); 207 women, 69 men; mean age 33 years	MBSR	Escitalopram	24 weeks	MBSR and escitalopram showed no significant differences in anxiety or depression outcomes	Mixed or inconclusive	MD: -1.6 (95% CI -3.4–0.3)
Siegfried Kasper, 2024	RCT (multi-centre, double-blind, randomised, placebo- and reference-controlled Phase III trial)	498	Adult patients (≥18 years) suffering from a major depressive episode of mild to moderate severity according to ICD-10 with a MADRS total score of 19–34 points	Silexan (80 mg) and Sertraline (50 mg)	Placebo	8 weeks	Silexan was superior to placebo and not different from Sertraline in reducing MADRS scores	Mixed or inconclusive	adjusted mean change in MADRS total score: -12.1 (95% CI -13.3–-11.0, p=< 0.01)
Andri Rennwald, 2026	Prospective longitudinal observational cohort study	32	Swiss adult primary care patients aged 18 to 75 years in stable remission who had taken antidepressants for at least six months and were planning to discontinue	Antidepressant dose reduction/discontinuation	NR	26 weeks (6 months)	Dose reductions below 75% of minimum effective dose associated with withdrawal symptom increases	Mixed or inconclusive	OR: 3.2 (95% CI 1.4–7.4)
Katarina Kopcalic, 2025	Systematic review and meta-analysis	NR	Adults with a primary diagnosis of moderate-to-severe generalised anxiety disorder (GAD) without serious medical comorbidities	Antidepressants	Placebo	4 weeks to 28 weeks	Antidepressants are more effective than placebo at improving treatment response in GAD	Mixed or inconclusive	RR: 1.41 (95% CI 1.29–1.55)
Judith Cukor, 2024	observational cohort	577	577 adult patients (≥18 years) with at least moderate depression (PHQ-9 ≥10) and anxiety (GAD-7 ≥8) who received antidepressant prescriptions at an urban academic medical center in New York City	Antidepressant treatment	NR	3 months (13 weeks)	Six distinct trajectories identified; fewer than half of patients showed symptom response	Mixed or inconclusive	RR: 1.02 (95% CI 1.0–1.05, p=0.041)
T. Jiménez Aparicio, 2024	Cross-sectional descriptive study	222	Adult patients referred for a first consultation in a mental health team for symptoms of anxiety and depression at the Hospital Clínico Universitario de Valladolid, Spain	Antidepressant treatment (with or without benzodiazepines)	NR	6 months	Up to 34% of patients abandoned antidepressant treatment within 6 months	Supports the study goal	NR
Hasan Mirzazadeh, 2025	RCT (double-blind, placebo-controlled, randomized clinical trial)	60	Adults aged 28-50 years diagnosed with OCD per DSM-V criteria with Y-BOCS scores ≥18, recruited from Namazi Hospital and Ibn Sina Polyclinic in Shiraz, Iran	Escitalopram plus memantine	Escitalopram plus placebo	16 weeks	Both groups significantly reduced OCD symptoms with no significant between-group difference	Mixed or inconclusive	Time X Group interaction F-statistic (Y-BOCS): 2.44 (p=0.12)
Francesco Monaco, 2025	Naturalistic observational study (non-randomized, 2×2 repeated-measures design)	45	Forty-five female inpatients with restrictive anorexia nervosa and comorbid OCD, aged ≥16 years, from a specialized residential eating disorders centre in Salerno, Italy	Lurasidone–fluvoxamine combination within structured inpatient rehabilitation	Structured inpatient rehabilitation without lurasidone–fluvoxamine	6 months	Improvements more plausibly attributed to global rehabilitation effects than specific pharmacological action	Mixed or inconclusive	NR
Jurriaan M J L Brouwer, 2024	Observational longitudinal cohort study (naturalistic)	928	Adults (18–65 years) with depressive and/or anxiety disorders from the NESDA cohort who used an antidepressant drug during 9-year follow-up, of North-European ancestry	Antidepressant use stratified by CYP2D6/CYP2C19 metabolizer status	NR	9 years	No significant associations between CYP metabolizer status and antidepressant switching or discontinuation	Mixed or inconclusive	OR: 1.34 (95% CI 0.89–2.02)
Toshiaki Kikuchi, 2024	Cross-sectional web-based survey	3376	Adults aged 18–59 years in Japan who self-reported MDD diagnosis and had been taking antidepressant medication for at least 3 months	Antidepressant medication	NR	NR	67.1% of MDD patients taking antidepressants self-reported symptoms of emotional blunting	Supports the study goal	prevalence: 67.1 (95% CI 65.5–68.7)
Sucharita Mandal, 2025	RCT (study protocol only; no results reported)	114	Adults aged 18–65 years diagnosed with OCD based on ICD-11 criteria, with Y-BOCS score ≥16, drug-naïve for at least seven days	Vortioxetine adjunctive to escitalopram	Placebo adjunctive to escitalopram	12 weeks	Protocol for RCT of vortioxetine as adjunctive treatment to escitalopram in OCD; no results	Mixed or inconclusive	NR
Amanda Hazel Dilmore, 2025	cross-sectional observational	666	Adults from the Tulsa-1000 and NeuroMAP CoBRE cohorts, including 164 healthy comparisons and 502 individuals diagnosed with anxiety disorder and/or MDD	Antidepressant use	No antidepressant use / healthy comparisons	NR	Antidepressant use associated with significant gut microbiota beta diversity differences	Mixed or inconclusive	NR
Luca Giacovelli, 2025	Retrospective observational study	10	Adults aged 18–65 years with treatment-resistant OCD (DSM-5) who had at least one inadequate SRI trial, treated at outpatient services in Milan, Italy	Brexpiprazole augmentation of SRI therapy	NR	12 weeks	Brexpiprazole augmentation of SRI resulted in 70% response rate in treatment-resistant OCD	Mixed or inconclusive	NR
Diana Dubrall, 2025	Observational (analysis of spontaneous adverse drug reaction reports)	8046	Adult psychiatric patients in Germany for whom spontaneous ADR reports related to antidepressants, antipsychotics, and mood stabilizers were submitted	Antidepressants, antipsychotics, and mood stabilizers	NR	NR	ADR reports from physicians, pharmacists, and consumers differ and complement each other	Mixed or inconclusive	NR
Rachael Frost, 2026	systematic review	NR	Adults diagnosed with MDD, mild-moderate depressive disorder, GAD, post-ischaemic stroke depression, multiple sclerosis with depressive symptoms, or unipolar depressive disorder taking antidepressants with herbal products	Herbal products adjunctive to antidepressants (SSRIs, SNRIs, TCAs)	Placebo adjunctive to antidepressants	NR	Adjunctive herbal products with antidepressants showed no significant increase in serious adverse events	Mixed or inconclusive	NR

Per-stage exclusion reasons are summarized below (PRISMA 2020 Item 16b).

Stage	Reason	Count
Title and abstract screening	Below Threshold	876

Included studies are stratified by study design below.

Study Type	Count
Randomized controlled trials	7
Systematic reviews and meta-analyses	6
Observational studies	14
Other study designs	3

3.2 Synthesis of Findings

Note on evidence hierarchy. This review's included set mixes primary studies (RCTs and/or observational designs) with secondary evidence (systematic reviews and/or meta-analyses). Reviewers should consider double-counting risk and weight primary and secondary evidence accordingly.

1. Direction-of-Effect Summary

Of the 30 included articles, 2 (6.7%) reported findings that directly support the study goal — that is, they provided evidence bearing on the association between SSRI or SNRI treatment and emotional blunting, reduced emotional authenticity, or subjective personality changes. The remaining 28 articles (93.3%) were classified as mixed or inconclusive with respect to the study goal, largely because they addressed adjacent clinical questions (treatment efficacy, adverse-event profiling, neuroimaging correlates, pharmacogenomics, or treatment discontinuation) without directly measuring emotional blunting or related subjective emotional outcomes.

Studies supporting the study goal (2 of 30, 6.7%):

Kikuchi and colleagues conducted a cross-sectional web-based survey of 3,376 adults with self-reported major depressive disorder in Japan who had been taking antidepressants for at least three months. They found that 67.1% of respondents endorsed symptoms of emotional blunting, with significant correlations between emotional blunting severity and depression, anxiety, social functioning, and quality-of-life measures. Reported significance: all p < 0.0001.

Jiménez Aparicio and colleagues described treatment adherence patterns in 222 adults with anxious-depressive syndrome, reporting that up to 34% abandoned antidepressant treatment within six months while 74% continued benzodiazepines — a pattern the authors interpreted as potentially reflecting differential tolerability or subjective experience of the two medication classes. This study did not report a formal significance metric.

Studies with mixed or inconclusive findings (28 of 30, 93.3%):

The large majority of retrieved articles did not directly assess emotional blunting, affective flattening, or subjective changes in emotional authenticity as a primary or secondary outcome. Instead, these studies addressed SSRI or SNRI efficacy, safety, or biological correlates in populations with depressive or anxiety-spectrum disorders, contributing contextual evidence about the therapeutic and adverse-effect landscape within which emotional blunting occurs but not measuring the phenomenon itself.

Reported significance metrics for these 28 studies were as follows. Cohen (2025, diagnostic test accuracy meta-analysis): PPV 86% (95% CI = 83–88%); sex difference in predictive accuracy β = −0.64, 95% CI = −1.12 to −0.16, p = 0.0089. Shukla (2025, prospective observational): p = 0.035 for association between tobacco use and multiple ADR occurrence. Cipriani (2026, RCT): P = .007. Mills (2024, Bayesian hierarchical meta-analysis): SSRI youth vs adults p = 0.129; SNRI youth vs adults p = 0.018; placebo youth vs adults p = 0.750; adult improvement from baseline by week 2 p < 0.001; youth improvement from baseline by week 2 p < 0.001. Otsubo (2025, RCT): P = 0.012. Van der Straten (2024, longitudinal nonrandomized controlled trial): p < 0.05 FWE-corrected at cluster level; symptom provocation insula cluster p(SVC) < 0.001; CBT cerebellar increase pFWE = 0.021; CBT insula/caudate increase during SST pFWE < 0.001. Pala Avan (2025, cross-sectional): p < 0.001. Han (2025, longitudinal observational): GRF corrected (voxel p < 0.001, cluster p < 0.05, two-tailed); Bonferroni correction applied for multiple comparisons. Kishi (2025, systematic review and meta-analysis): p < 0.01. Cohen (2025, IPDMA): p < 0.0001. Patel (2026, prospective observational cohort): p < 0.001 for between-group depression outcome (ANCOVA); P = 0.59 for between-group anxiety outcome (not significant). Xue (2026, prospective cohort): all p > 0.05 for prediction of treatment response; p < 0.001 for baseline group differences in oxytocin levels and genotype. Hu (2024, secondary analysis of RCT): no significant between-group differences were detected at week 8 (primary end point); adverse events significantly differed (P < .001). Kasper (2024, RCT): Silexan vs placebo p < 0.01; Sertraline vs placebo p < 0.01; responder and remission rates in both active groups vs placebo p < 0.05. Rennwald (2026, prospective longitudinal observational cohort): p < 0.001. Kopcalic (2025, systematic review and meta-analysis): 95% CI 1.29 to 1.55; high-certainty evidence. Cukor (2024, observational cohort): p = 0.041 for older age associated with moderate nonresponse trajectory; p = 0.021 for lower baseline PHQ-9 associated with moderate nonresponse trajectory. Mirzazadeh (2025, RCT): Time × Group interaction for Y-BOCS: F = 2.44, P = 0.12 (not significant between groups); Time management BDEFS between groups: P = 0.03. Monaco (2025, naturalistic observational): BF10 values for time effects on primary outcomes ranged from ~3.68 × 10³ to ~1.68 × 10¹² (very strong/decisive evidence); group × time interaction BF < 1 (no evidence exceeding anecdotal strength); MCAR assumption not refuted (p = 0.192). Brouwer (2024, observational longitudinal cohort): no significant associations (all 95% confidence intervals crossed 1.0). Dilmore (2025, cross-sectional observational): antidepressant use associated with beta diversity: UWUF, WUF, RPCA p = 0.001; MDD diagnosis: UWUF p = 0.001; ANXD diagnosis not significant after multiple test correction. Giacovelli (2025, retrospective observational): P < 0.01 (paired t-test for mean Y-BOCS total score at week 12 vs baseline); P < 0.001 (repeated-measures ANOVA).

Six of the 28 mixed-or-inconclusive studies did not report a significance metric: Rai (2024, RCT protocol with no results), Orsolini (2024, case report), Frost (2026, systematic review), Kohat (2026, case report), Mandal (2025, RCT protocol with no results), and Dubrall (2025, observational analysis of spontaneous ADR reports).

2. Consistency vs. Contradiction Analysis

The included studies rarely addressed the same outcome using comparable designs, populations, and measurement instruments, which limits the extent to which true contradictions can be identified. Where findings appear to diverge, the differences are more often attributable to heterogeneity in study purpose and design than to genuine empirical conflict.

Emotional blunting: sparse and methodologically isolated evidence

Only one study directly measured emotional blunting prevalence in antidepressant-treated adults. Kikuchi and colleagues reported that 67.1% of Japanese adults with self-reported major depressive disorder who had been taking antidepressants for at least three months endorsed symptoms of emotional blunting on the Oxford Depression Questionnaire, with blunting severity correlating significantly with depression, anxiety, functional impairment, and reduced quality of life. By contrast, the prospective observational study by Shukla and colleagues in drug-naïve adults with generalized anxiety disorder found no reports of emotional blunting or affective flattening during six weeks of SSRI monotherapy, though the sample was small (n = 45) and adverse events were captured through spontaneous self-report rather than a validated emotional-blunting instrument. These two findings are not necessarily contradictory: the populations differed markedly in diagnosis, treatment duration, cultural context, and ascertainment method. Kikuchi's cross-sectional survey recruited patients with months of antidepressant exposure and used a purpose-built emotional-blunting scale, whereas Shukla's study followed newly treated patients over a short interval using general adverse-event monitoring that may lack sensitivity for subjective emotional changes. The discrepancy highlights that the likelihood of detecting emotional blunting depends heavily on whether investigators specifically ask about it and how long patients have been treated.

The neuroimaging study by Han and colleagues offers a different vantage point. After four weeks of paroxetine in drug-naïve OCD patients, enhanced bottom-up striatal-to-cortical connectivity was observed, and the authors interpreted this as evidence that paroxetine may normalize aberrant circuitry rather than produce emotional blunting. This interpretation, however, rests on a neurobiological inference rather than on patient-reported emotional experience, and the absence of a validated emotional-blunting measure in that study means it cannot directly confirm or refute the prevalence figure reported by Kikuchi.

Antidepressant efficacy in anxiety and OCD: broadly consistent, with expected variation in effect size

Across the studies that examined SSRI or SNRI efficacy in anxiety-spectrum and obsessive-compulsive disorders, the direction of effect was consistently favorable relative to placebo, though the magnitude varied. The large Cochrane-style meta-analysis by Kopcalic and colleagues found antidepressants superior to placebo for generalized anxiety disorder treatment response (RR 1.41, 95% CI 1.29–1.55; NNTB 7), rated as high-certainty evidence. The individual-patient-data meta-analysis by Cohen and colleagues similarly demonstrated SSRI superiority over placebo in OCD, albeit with a small effect size (Hedges' g 0.33; mean difference 2.65 Y-BOCS points; NNT 7). Otsubo and colleagues confirmed venlafaxine ER's superiority over placebo in Japanese adults with GAD (LS mean difference −1.9, p = 0.012). The Bayesian hierarchical meta-analysis by Mills and colleagues found that both SSRIs and SNRIs produced early improvement in adults with anxiety disorders, with SNRIs showing a larger response in adults than in youth.

These efficacy findings are internally consistent. The modest absolute effect sizes reported across studies are compatible with one another and with the broader antidepressant literature. No included study found antidepressants inferior to placebo for the primary anxiety or OCD indication.

Treatment discontinuation and adherence: convergent signals from different angles

Several studies touched on treatment discontinuation, though from different perspectives. Jiménez Aparicio and colleagues found that 34% of patients with anxious-depressive syndrome abandoned antidepressant treatment within six months, while 74% continued benzodiazepines — a pattern the authors interpreted as reflecting differential tolerability or patient preference. Cukor and colleagues identified six distinct symptom trajectories during antidepressant treatment, with fewer than half of patients meeting response criteria for either depression or anxiety over 13 weeks. Rennwald and colleagues documented that dose reductions below 75% of the minimum effective dose triggered clinically relevant withdrawal symptoms in roughly one-third of tapering intervals. The Cipriani trial found that a clinical decision-support tool reduced all-cause treatment discontinuation compared with usual care (adjusted RR 0.62, 95% CI 0.44–0.88).

These findings are not contradictory but illuminate different facets of the same problem. High early discontinuation rates, heterogeneous symptom trajectories, and withdrawal phenomena during tapering all point toward a complex treatment-adherence landscape. None of these studies, however, specifically attributed discontinuation to emotional blunting or subjective personality change, leaving a gap in the evidence chain connecting emotional side effects to treatment non-persistence.

Adverse-effect profiles: consistent in type, variable in ascertainment

Shukla's prospective study found predominantly mild, self-limiting adverse reactions to SSRIs in GAD (88.6% mild; 50% central nervous system, 33% gastrointestinal), with no severe events. Hu and colleagues, in a secondary analysis of an RCT comparing mindfulness-based stress reduction with escitalopram for anxiety disorders, reported that escitalopram recipients experienced significantly more adverse events overall (78.6% vs. 15.4%), though between-group differences in anxiety, depression, and quality-of-life outcomes were small and non-significant. The spontaneous adverse-event reporting analysis by Dubrall and colleagues found that consumers were more likely than physicians or pharmacists to report sexual dysfunction and that reporter type influenced which adverse reactions were captured. These observations are consistent with one another: the type and frequency of detected adverse effects depend on who is asked, how they are asked, and what instruments are used. The absence of emotional blunting from most adverse-event inventories likely reflects the use of general safety monitoring rather than targeted emotional-outcome assessment.

Brain-level and biological studies: complementary rather than conflicting

Van der Straten and colleagues found that SSRIs and CBT produced largely opposite patterns of brain-activity change in OCD patients — SSRIs decreased activity broadly while CBT increased it — yet both treatments reduced insula activation during symptom provocation. Han and colleagues' finding of enhanced striatal-to-cortical connectivity after paroxetine addresses a different circuit and a different imaging paradigm. Dilmore and colleagues reported that antidepressant use was associated with gut microbiota differences exceeding those attributable to psychiatric diagnosis alone. Xue and colleagues found that baseline oxytocin levels and OXTR genotype were associated with GAD disease state but did not predict escitalopram response. These biological studies address distinct mechanisms and do not contradict one another; they collectively suggest that serotonergic antidepressants exert measurable neurobiological and systemic effects, but none of these studies linked such effects to patient-reported emotional blunting or subjective personality change.

Sources of heterogeneity across the evidence set

The most prominent sources of apparent inconsistency across the included studies are differences in (a) the outcome measured — most studies assessed symptom reduction or adverse events rather than emotional blunting per se; (b) the population studied — diagnoses ranged from major depression to OCD to GAD to anorexia nervosa with comorbid OCD, and settings spanned community surveys, tertiary care, and regulatory trial databases; (c) the duration of observation — follow-up ranged from four weeks to nine years, and emotional blunting may emerge or become salient only after sustained treatment; and (d) the method of ascertainment — validated emotional-blunting instruments were used in only one study, while others relied on general adverse-event monitoring or did not assess emotional outcomes at all. These design-level differences mean that the absence of emotional-blunting findings in most studies cannot be interpreted as evidence against the phenomenon; it more likely reflects the fact that most investigators were not looking for it.

3. Study Limitations (Aggregated)

The included studies carried a range of methodological constraints that bear on the confidence with which their findings can be applied to the question of whether SSRI or SNRI treatment is associated with emotional blunting or subjective personality changes. The most frequently acknowledged limitation was small sample size or insufficient statistical power: at least 10 studies explicitly noted that their sample was too small to detect clinically meaningful effects, to support planned subgroup analyses, or to generalize findings to broader populations. Several of these were single-case reports or pilot-scale investigations with fewer than 50 participants, and one trial enrolled only 32 of a planned 400 participants owing to recruitment difficulties.

Absence of randomization or blinding was a recurring concern. Multiple observational and naturalistic studies acknowledged that non-random treatment allocation introduced potential confounding by indication — patients who received combined pharmacotherapy and psychotherapy, for instance, tended to have higher baseline symptom severity. One longitudinal neuroimaging study noted that the lack of randomization between SSRI and cognitive-behavioral therapy arms may have biased between-group comparisons. Even among the randomized trials, one acknowledged that its open-label design limited the validity of its results.

Restricted generalizability was cited across study designs and populations. Several trials excluded patients with medical comorbidities, substance use, or treatment-resistant illness, and authors recognized that these exclusions narrow the applicability of findings to the heterogeneous populations seen in routine clinical practice. Single-center recruitment was flagged in at least five studies. Population-specific constraints were also noted: one pharmacogenomic study limited its conclusions to individuals of Han Chinese ethnicity, another to a rural Indian tertiary-care setting, and a cross-sectional survey relied entirely on self-reported diagnoses of major depressive disorder without physician confirmation.

Short follow-up duration was a common concern, particularly given that emotional blunting may emerge or evolve over longer treatment courses. Most treatment trials lasted 6 to 16 weeks, and several authors explicitly stated that their observation windows were too brief to capture delayed adverse effects or to evaluate whether initial benefits were sustained. Only a handful of studies followed participants beyond 24 weeks.

Reliance on self-report instruments without objective or neurobiological validation was noted in multiple studies. One naturalistic study of anorexia nervosa with comorbid OCD acknowledged that its exclusive use of self-report questionnaires, combined with missing data rates exceeding 60% on some variables, introduced substantial uncertainty. The cross-sectional survey that directly measured emotional blunting prevalence similarly relied on patient self-report and could not confirm diagnoses independently. A study of antidepressant-associated bruxism noted the absence of polysomnographic confirmation.

Missing data and attrition posed interpretive challenges in several investigations. One cohort study reported missing data ranging from 20% to over 60% across key variables and relied on multiple imputation. A clinical decision-support trial acknowledged that a large amount of missing data limited the validity of its findings. The use of last-observation-carried-forward imputation — which assumes data are missing completely at random — was flagged as a potential source of bias in one individual-patient-data meta-analysis.

Several studies acknowledged an inability to disentangle medication effects from other influences. One observational cohort studying symptom trajectories during antidepressant treatment noted that, without a placebo arm, observed improvements could not be distinguished from regression to the mean. A naturalistic inpatient study concluded that temporal improvements were more plausibly attributable to the structured rehabilitation program than to the specific pharmacological combination under study. The absence of active comparators, dose-response data, or concurrent assessment of polypharmacy was noted in multiple reports.

Notably, the cross-sectional design of the study most directly relevant to emotional blunting prevalence precluded causal inference about the direction of the association between antidepressant use and blunting symptoms; the authors acknowledged that patients' attitudes toward emotional blunting likely varied with disease stage and treatment course. Two included records were study protocols without reported results, contributing no outcome data to the evidence base. Across the set, no study was specifically designed with emotional blunting as a primary endpoint measured by validated instruments alongside a controlled comparator, which limits the strength of conclusions that can be drawn about the relationship between SSRI or SNRI treatment and changes in emotional experience.

4. Population Heterogeneity

The included studies enrolled populations that span several, but not all, of the diagnostic categories specified in the review's eligibility criteria. Major depressive disorder and generalized anxiety disorder were the most frequently represented conditions, followed by obsessive-compulsive disorder; panic disorder and social anxiety disorder appeared primarily within mixed-diagnosis cohorts or as co-eligible conditions in broader anxiety-disorder samples. No included study focused exclusively on persistent depressive disorder (dysthymia) as the primary diagnosis, and only one large naturalistic cohort included participants with dysthymia alongside other depressive and anxiety diagnoses. This uneven diagnostic coverage means that conclusions about emotional blunting and related subjective emotional changes are most directly informed by evidence from MDD and GAD populations, with OCD contributing data on symptom trajectories and treatment response but rarely measuring emotional blunting as an outcome.

Diagnostic Subgroups

Major depressive disorder. The single study that directly measured emotional blunting prevalence was conducted in adults with self-reported MDD in Japan and found that 67.1% of antidepressant-treated patients endorsed symptoms of emotional blunting, with severity correlating with depression, anxiety, functional impairment, and quality-of-life measures. A separate observational cohort from New York identified six distinct symptom trajectories during antidepressant treatment, with fewer than half of patients meeting response criteria for either depressive or anxiety symptoms over thirteen weeks; older age and lower baseline depression severity were associated with nonresponse. A cross-sectional study from Spain reported that up to 34% of patients with anxious-depressive syndrome abandoned antidepressant treatment within six months, though the reasons for discontinuation were not systematically linked to emotional side effects. These MDD-focused findings collectively suggest that emotional blunting is common in antidepressant-treated depression, but the cross-sectional designs preclude distinguishing medication-attributable emotional changes from residual illness effects.

Generalized anxiety disorder. Several studies enrolled adults with GAD, yet none directly assessed emotional blunting or affective flattening as a primary or secondary outcome. One prospective observational study of drug-naïve GAD patients treated with SSRIs reported no cases of emotional blunting among 45 participants over six weeks, though the authors acknowledged that the short follow-up and small sample may have limited detection. A placebo-controlled trial of venlafaxine extended-release in Japanese outpatients with GAD demonstrated efficacy on anxiety symptom scales without reporting emotional side effects. A Cochrane-level systematic review and meta-analysis encompassing over 12,000 participants confirmed that antidepressants are more effective than placebo for GAD treatment response, with a number needed to treat of seven, but noted that more participants in antidepressant groups discontinued due to adverse effects — the nature of those adverse effects was not disaggregated to identify emotional blunting specifically. A secondary analysis comparing mindfulness-based stress reduction with escitalopram in adults with primary anxiety disorders found no meaningful between-group differences in anxiety, depression, or quality-of-life outcomes, though escitalopram recipients experienced significantly more adverse events overall.

Obsessive-compulsive disorder. OCD was the focus of the largest cluster of included studies, yet these investigations centered on symptom reduction, treatment response prediction, and neurobiological mechanisms rather than on emotional blunting per se. An individual-participant-data meta-analysis of over 2,300 adults found SSRIs superior to placebo for OCD with a small effect size, and a diagnostic-accuracy meta-analysis showed that early non-improvement predicted later nonresponse with high positive predictive value. A neuroimaging study of paroxetine-treated drug-naïve OCD patients reported changes in striatal-cortical connectivity and explicitly noted that these changes suggested symptom normalization rather than emotional blunting. A meta-analysis of maintenance antidepressant treatment in stabilized OCD patients found that continuation reduced relapse risk substantially (relative risk 0.53) but did not report on emotional side effects during maintenance. None of the OCD studies used validated instruments for emotional blunting.

Age and Developmental Stage

Most included studies enrolled adults broadly defined (ages 18–65 or 18–75), and only one meta-analysis explicitly compared youth and adult responses to SSRIs and SNRIs for anxiety disorders. That analysis found comparable SSRI response across age groups but greater SNRI response in adults than in youth, suggesting a possible developmental modulation of serotonergic-noradrenergic treatment effects. No study examined whether emotional blunting prevalence or severity varied by age within adult samples, although one observational cohort noted that older age was associated with depressive symptom nonresponse during antidepressant treatment.

Severity at Baseline

Baseline symptom severity was inconsistently reported and rarely examined as a moderator of emotional outcomes. The individual-participant-data meta-analysis of SSRIs for OCD found treatment effects independent of baseline patient characteristics, including severity. In the MDD symptom-trajectory study, lower baseline depression severity was associated with membership in a nonresponse trajectory, a pattern that could reflect floor effects or differential sensitivity to treatment. The Japanese emotional-blunting survey did not stratify prevalence by depression severity, limiting the ability to determine whether more severely ill patients experience more or less emotional blunting during treatment.

Sex and Gender

Sex-stratified findings were sparse. The OCD diagnostic-accuracy meta-analysis identified a statistically significant sex difference in predictive accuracy for early non-improvement (β = −0.64, p = 0.009), suggesting that the predictive model performed differently in men and women, though the clinical implications for emotional outcomes were not explored. The naturalistic study of anorexia nervosa with comorbid OCD enrolled only female participants, precluding sex comparisons. No study examined sex as a moderator of emotional blunting.

Treatment Setting

Studies were conducted across outpatient, inpatient, and community settings, but setting was not examined as a moderator of emotional outcomes in any included study. The inpatient study of women with anorexia nervosa and comorbid OCD attributed observed improvements to the global rehabilitation milieu rather than to specific pharmacological effects, raising the possibility that treatment setting confounds the interpretation of emotional changes during antidepressant use. Most remaining studies were outpatient-based, and the two protocol-only publications described outpatient or community recruitment without reporting results.

Geographic and Cultural Context

The included studies were conducted across diverse geographic settings — Japan, India, Brazil, Canada, the United Kingdom, the Netherlands, Switzerland, Spain, Germany, Iran, Italy, China, and the United States — yet only the Japanese cross-sectional survey directly measured emotional blunting prevalence. The Indian GAD safety study noted that the absence of reported sexual dysfunction likely reflected cultural underreporting, raising the analogous concern that emotional blunting may also be underreported in settings where patients are less accustomed to volunteering subjective emotional complaints. The German pharmacovigilance analysis found that consumers, compared with physicians and pharmacists, more commonly reported certain psychiatric adverse drug reactions, suggesting that reporter type and cultural context shape which side effects enter the evidence base.

Gaps Relative to the Prespecified PICO

Several population segments specified in the eligibility criteria are poorly represented or absent. No included study focused on adults with social anxiety disorder or panic disorder as the sole diagnosis, though these conditions appeared within mixed anxiety-disorder samples. Persistent depressive disorder received no dedicated investigation. Adults older than 65 years were explicitly excluded from most studies, leaving the relationship between SSRI or SNRI treatment and emotional blunting in older adults unexamined. Patients with substantial medical or psychiatric comorbidities were excluded from nearly all trials and cohort studies, and the Cochrane-level review of antidepressants for GAD acknowledged that its findings cannot be extended to patients with co-occurring medical conditions. The two study protocols — one examining sertraline in autistic adults with anxiety, the other examining vortioxetine augmentation of escitalopram in OCD — may eventually contribute relevant data but have not yet reported results.

5. Effect Magnitude Language (Aggregated)

The language used to characterize effect magnitude across the included studies was heterogeneous, reflecting the diversity of study designs, populations, and outcomes. Several patterns are worth noting.

Among the meta-analytic and large-scale studies, effect sizes were described in quantitative but varied terms. One individual-patient-data meta-analysis of SSRIs for OCD reported "a small effect size" (Hedges' g = 0.33), corresponding to a mean difference of 2.65 Y-BOCS points and a number needed to treat of seven. A Cochrane-style systematic review of antidepressants for generalized anxiety disorder described the treatment benefit as a number needed to treat for an additional beneficial outcome of 7 (95% CI 5 to 9). A meta-analysis of antidepressant maintenance in OCD characterized the benefit as a relative risk of 0.53 (95% CI 0.42–0.68) for relapse. A diagnostic-accuracy meta-analysis framed its findings in terms of predictive values rather than treatment effects, reporting a positive predictive value of 86% (95% CI 83–88%) for early non-improvement predicting later nonresponse. These descriptions, while precise, are not directly commensurable with one another.

Randomized trials tended to report least-squares mean differences or between-group contrasts. The venlafaxine trial in Japanese adults with GAD reported a least-squares mean difference of −1.9 (95% CI −3.4 to −0.4) on the HAM-A. The comparison of mindfulness-based stress reduction with escitalopram described between-group differences as corresponding to "small effect sizes (Cohen d ≤ 0.20)," interpreted as a lack of clinically meaningful separation. A trial of Silexan versus sertraline versus placebo reported adjusted mean MADRS reductions of 12.1, 12.6, and 9.95 points, respectively, characterizing the active-treatment effects as "clinically relevant" without specifying a standardized effect size. The escitalopram-plus-memantine trial in OCD reported substantial within-group Y-BOCS reductions in both arms but no significant between-group difference (interaction F = 2.44, p = 0.12).

Observational and cohort studies used still different framing. The antidepressant-discontinuation cohort described withdrawal symptom increases with an odds ratio of 3.2 (95% CI 1.4–7.4) for dose reductions below 75% of the minimum effective dose. The bruxism cross-sectional study reported an odds ratio of 5.928 (95% CI 3.708–9.478) for antidepressant use predicting bruxism. The Japanese cross-sectional survey of emotional blunting reported a self-reported prevalence of 67.1% and correlation coefficients between emotional blunting severity and depression, anxiety, social functioning, and quality-of-life measures ranging from 0.51 to 0.67 in absolute value. The antidepressant-adherence study described its finding in proportional terms — up to 34% of patients abandoning treatment within six months — without a formal effect-size metric. A naturalistic study of anorexia nervosa with comorbid OCD reported Bayes factors for time effects ranging from approximately 10³ to 10¹², interpreted as "very strong" to "decisive" evidence for temporal improvement, but found no evidence for group-by-time interactions.

Several studies did not report effect magnitudes in conventional terms. The gut-microbiota study noted that antidepressant use had "a larger effect size than neuropsychiatric disorder diagnosis" on beta diversity but did not quantify this comparison numerically. The pharmacogenomic cohort study reported that it was underpowered (calculated power 0.06–0.29) and found no significant associations. Two protocol-only publications and the spontaneous adverse-event reporting analysis did not report effect magnitudes because no efficacy outcomes were available.

Overall, the language of effect magnitude ranged from standardized metrics (Cohen's d, Hedges' g, partial η²) through absolute and relative risk measures (number needed to treat, relative risk, odds ratios, absolute risk reduction) to descriptive proportions and Bayes factors. No single metric predominated, and direct comparison of effect sizes across studies is not straightforward given the different outcome domains and analytic frameworks.

Significance Metrics Reported

Of the 30 included articles, 20 reported at least one explicit p-value. Fourteen reported confidence intervals (whether for effect estimates, odds ratios, relative risks, or predictive values). Three articles stated that findings were statistically significant or non-significant without providing a specific numeric threshold beyond the p-value or confidence interval already counted. Seven articles — comprising two study protocols without results, one systematic review of herbal adjuncts, one case report, one adverse-event reporting analysis, one treatment-adherence descriptive study, and one Bayesian meta-analysis reporting trajectory-level results — did not report a conventional significance metric or reported null values for this field. Several studies reported both p-values and confidence intervals, so these categories are not mutually exclusive.

6. Study Design Profile

The 30 included articles encompass a broad range of study designs, and this heterogeneity has direct implications for the strength and type of inferences that can be drawn about the relationship between SSRI or SNRI treatment and emotional blunting or subjective emotional changes in adults with depressive or anxiety disorders.

Randomized controlled trials constitute a modest share of the evidence base. Seven records overall were classified as randomized controlled trials: four completed RCTs, two protocol-only publications without results, and one secondary analysis of a completed trial. The four completed RCTs are a Japanese trial of venlafaxine extended-release versus placebo in generalized anxiety disorder, a multi-centre trial comparing Silexan, sertraline, and placebo in mild-to-moderate major depressive episodes, a double-blind trial of escitalopram with or without memantine augmentation in OCD, and a pragmatic trial evaluating a clinical decision-support tool for personalized antidepressant selection in major depressive disorder. The two RCT protocols — one for sertraline in anxious adults with autism and one for vortioxetine adjunctive to escitalopram in OCD — that have not yet reported outcome data and therefore cannot contribute to the synthesis of findings. A secondary analysis of a randomized trial comparing mindfulness-based stress reduction with escitalopram in primary anxiety disorders also appears. Collectively, the completed trials address symptom reduction and treatment acceptability but were not designed to measure emotional blunting as a primary or secondary outcome, limiting their direct relevance to the central question of this review.

Systematic reviews and meta-analyses represent the largest analytic studies in the set. These include a Cochrane-style meta-analysis of antidepressants versus placebo for generalized anxiety disorder (encompassing over 12 000 participants), two individual-participant-data meta-analyses of SSRIs in OCD (one focused on treatment efficacy, the other on early prediction of nonresponse), a Bayesian hierarchical meta-analysis of SSRI and SNRI response trajectories in anxiety disorders, a pairwise meta-analysis of antidepressant maintenance treatment in OCD, and a systematic review of herbal product–antidepressant interactions. While these syntheses draw on large pooled samples and provide high-certainty evidence for efficacy and relapse prevention, none was designed to capture emotional blunting, reduced emotional authenticity, or subjective personality change as outcomes.

Observational designs — prospective cohorts, cross-sectional surveys, longitudinal studies, and naturalistic comparisons — make up the remainder. A cross-sectional web-based survey of over 3 300 Japanese adults with self-reported major depressive disorder is the only study that directly measured emotional blunting prevalence using a validated instrument (the Oxford Depression Questionnaire), finding that roughly two thirds of antidepressant-treated respondents endorsed emotional blunting symptoms. Other observational studies examined symptom trajectories during antidepressant treatment, pharmacogenomic predictors of antidepressant switching and discontinuation, brain connectivity changes after paroxetine in OCD, antidepressant-associated bruxism, gut microbiota differences associated with antidepressant use, antidepressant discontinuation and withdrawal symptoms, treatment adherence patterns in anxious-depressive patients, and adverse drug reaction reporting patterns. A prospective observational study of SSRI safety in drug-naïve patients with generalized anxiety disorder explicitly noted the absence of emotional blunting among recorded adverse events, though the small sample and six-week follow-up constrain interpretation. Two case reports — one describing sertraline-induced bruxism and one describing lurasidone augmentation of fluoxetine in OCD — and a small retrospective series of brexpiprazole augmentation in treatment-resistant OCD round out the evidence.

Several design-level features shape how this body of evidence should be read. First, the overwhelming majority of included studies were designed to evaluate symptom reduction, treatment response prediction, or pharmacological safety — not to assess emotional blunting or changes in emotional authenticity. The single study that directly measured emotional blunting prevalence used a cross-sectional, self-report design without physician-confirmed diagnoses, precluding causal attribution. Second, the absence of any controlled trial that randomized participants and then prospectively measured emotional blunting as a prespecified outcome means that the evidence cannot distinguish medication-induced emotional restriction from residual symptoms of the underlying depressive or anxiety disorder. Third, the two protocol-only publications contribute no outcome data. Fourth, the case reports and small open-label series, while clinically informative, carry inherent limitations of selection bias and absence of comparator groups. Finally, the neuroimaging study of paroxetine in OCD offers mechanistic data on striatal–cortical connectivity changes but was not powered or designed to link those changes to patients' subjective emotional experience.

Taken together, the design profile of this evidence set is weighted toward efficacy and safety questions rather than the experiential and qualitative dimensions of antidepressant treatment that the study goal addresses. The single directly relevant prevalence estimate comes from a cross-sectional survey, and no included study employed a controlled, prospective design with emotional blunting as a primary endpoint. This imbalance limits the certainty of any conclusions about the magnitude, frequency, or causal attribution of SSRI- or SNRI-related emotional blunting and underscores the need for purpose-designed prospective research on this outcome.

7. Recency of Evidence

The evidence base underpinning this review is notably current. All 30 included articles were published between 2024 and 2026: 12 appeared in 2024, 12 in 2025, and 6 carry a 2026 publication year. No study predates 2024. This concentration within a narrow and recent window reflects the search strategy's temporal scope and indicates that the findings synthesized here draw on the most contemporary literature available. A practical consequence is that longer-term follow-up data and replication of newer findings — particularly those from the two protocol-only publications and from small pilot or case-level reports — remain forthcoming. The absence of older landmark studies in the included set also means that foundational work on SSRI- and SNRI-associated emotional blunting published before 2024 is not directly represented, which the reader should bear in mind when weighing the breadth of the evidence captured.

8. Evidence Gaps

8a. Gaps Identified from Article Text

Several included studies explicitly acknowledged limitations that, taken together, delineate areas where the evidence base remains thin.

The large individual-participant-data meta-analysis of SSRIs for OCD (Cohen et al., 2025) noted that patient-reported outcome measures and quality-of-life data were wholly unavailable across the regulatory trial submissions it analyzed, and that adverse-event data — including emotional or affective side effects — could not be examined. The companion diagnostic-accuracy meta-analysis by the same group similarly noted that its placebo-controlled setting may not reflect real-world clinical practice and that functional or quality-of-life scales were not encompassed. The Cochrane-level systematic review of antidepressants for generalized anxiety disorder (Kopcalic et al., 2025) identified important gaps in the literature on antidepressants for GAD and called for greater transparency in methodology and outcome reporting, as well as inclusion of patients with medical comorbidities — a population systematically excluded from the trials it synthesized.

The prospective observational study of SSRI safety in GAD (Shukla et al., 2025) acknowledged that its small sample (n = 45), single-center design, exclusion of elderly patients and those with comorbidities, and reliance on self-reported adverse reactions without systematic laboratory monitoring may have led to underestimation of subclinical effects. The pharmacogenomic cohort study (Brouwer et al., 2024) recognized that it was underpowered to detect associations between CYP2D6 or CYP2C19 metabolizer status and antidepressant switching or discontinuation, with calculated power as low as 0.06 for some comparisons, and that most patients used SSRIs whose broad therapeutic indices make concentration–effect relationships less clear than for tricyclic antidepressants.

The cross-sectional survey of emotional blunting in Japanese patients with major depressive disorder (Kikuchi et al., 2024) — the only included study that directly measured emotional blunting prevalence — noted that diagnoses were self-reported rather than physician-confirmed, that the effect of antidepressant class on emotional blunting was not analyzed, and that the cross-sectional design could not capture how attitudes toward emotional blunting change across disease stages and treatment courses.

The antidepressant discontinuation study (Rennwald et al., 2026) achieved only 32 of its planned 400 participants, precluding the originally intended analyses of predictor and moderator variables of withdrawal reactions. The naturalistic study of lurasidone–fluvoxamine in anorexia nervosa with comorbid OCD (Monaco et al., 2025) acknowledged missing data ranging from 20% to over 60% across key variables, an all-female sample, and the absence of neurobiological or behavioral validation of self-report instruments. The brexpiprazole augmentation study (Giacovelli et al., 2025) noted its small sample (n = 10), retrospective open design, absence of a control group, and inability to analyze confounding variables.

Two included records were study protocols without results (Rai et al., 2024; Mandal et al., 2025), underscoring that trials designed to address some of these questions — sertraline for anxiety in autistic adults and vortioxetine augmentation of escitalopram in OCD — have not yet reported outcome data.

8b. Gaps Identified by Independent Analysis

Comparison of the prespecified PICO against the content of the retrieved evidence set reveals several notable absences.

Within the retrieved evidence set, no studies used a validated instrument specifically designed to measure emotional blunting (such as the Oxford Depression Questionnaire) as a prospective primary or secondary outcome in a controlled trial of SSRIs or SNRIs. The single study that employed the ODQ (Kikuchi et al., 2024) was a cross-sectional survey without a comparator arm. Whether this represents a gap in the broader literature or a limitation of this search strategy is beyond the scope of this review.

Within the retrieved evidence set, no studies directly compared the emotional side-effect profiles of SSRIs against SNRIs, or of serotonergic antidepressants against non-serotonergic antidepressants, with respect to emotional blunting or reduced emotional responsiveness. Whether this represents a gap in the broader literature or a limitation of this search strategy is beyond the scope of this review.

Within the retrieved evidence set, no studies examined subjective loss of emotional authenticity, perceived personality change, or reduced sense of self as a defined outcome of SSRI or SNRI treatment. Whether this represents a gap in the broader literature or a limitation of this search strategy is beyond the scope of this review.

Within the retrieved evidence set, no studies examined the impact of SSRI or SNRI treatment on creativity, motivation, or emotional empathy using standardized measures. Whether this represents a gap in the broader literature or a limitation of this search strategy is beyond the scope of this review.

Within the retrieved evidence set, no studies examined relationship satisfaction or interpersonal functioning as outcomes of antidepressant-associated emotional changes. Whether this represents a gap in the broader literature or a limitation of this search strategy is beyond the scope of this review.

Within the retrieved evidence set, no studies reported antidepressant discontinuation rates attributable specifically to emotional blunting or affective flattening as a reason for stopping treatment. The discontinuation data that were available (Jiménez Aparicio et al., 2024; Rennwald et al., 2026; Brouwer et al., 2024) addressed all-cause discontinuation or withdrawal symptoms without isolating emotional side effects as a driver. Whether this represents a gap in the broader literature or a limitation of this search strategy is beyond the scope of this review.

Within the retrieved evidence set, no studies employed a psychotherapy-only comparator arm to distinguish medication-attributable emotional blunting from residual affective symptoms of the underlying disorder. The single study comparing CBT and SSRI treatment in OCD (van der Straten et al., 2024) examined neuroimaging changes rather than subjective emotional experience. Whether this represents a gap in the broader literature or a limitation of this search strategy is beyond the scope of this review.

Within the retrieved evidence set, no studies examined dose–response relationships between SSRI or SNRI dosage and the severity of emotional blunting or affective flattening. Whether this represents a gap in the broader literature or a limitation of this search strategy is beyond the scope of this review.

With respect to population coverage, the retrieved studies were concentrated in OCD and GAD populations. Within the retrieved evidence set, no studies specifically examined emotional blunting in adults with social anxiety disorder, panic disorder, or persistent depressive disorder as the primary diagnosis. Whether this represents a gap in the broader literature or a limitation of this search strategy is beyond the scope of this review.

Geographically, the retrieved studies drew participants predominantly from East Asia (Japan, China), South Asia (India), Europe (the Netherlands, Switzerland, Germany, Spain, Italy, United Kingdom), and the Americas (United States, Brazil). Within the retrieved evidence set, no studies were conducted in sub-Saharan Africa, the Middle East (apart from Iran), or Oceania. Whether this represents a gap in the broader literature or a limitation of this search strategy is beyond the scope of this review.

With respect to follow-up duration, most included studies observed patients for 16 weeks or less. Only the pharmacogenomic cohort (Brouwer et al., 2024) followed participants for as long as nine years. Within the retrieved evidence set, no controlled studies examined the emergence, persistence, or resolution of emotional blunting over treatment durations exceeding six months. Whether this represents a gap in the broader literature or a limitation of this search strategy is beyond the scope of this review.

Finally, the retrieved set contained no randomized controlled trials in which emotional blunting or reduced emotional responsiveness served as a prespecified primary outcome. The evidence bearing on the review's central question is therefore drawn entirely from cross-sectional surveys, observational designs, and studies whose primary endpoints addressed symptom reduction or neuroimaging rather than subjective emotional experience. Whether this represents a gap in the broader literature or a limitation of this search strategy is beyond the scope of this review.

9. Clinical Implications

The evidence assembled in this review offers limited but clinically relevant guidance for practitioners prescribing SSRIs or SNRIs to adults with depressive or anxiety-spectrum disorders. One large cross-sectional survey found that roughly two-thirds of antidepressant-treated patients with major depressive disorder self-reported symptoms consistent with emotional blunting, and that the severity of these symptoms correlated with worse depression, anxiety, functional impairment, and quality of life. This prevalence estimate, however, rests on unconfirmed self-reported diagnoses and a single cultural context, and no controlled trial in the included set was designed with emotional blunting as a primary efficacy or safety endpoint. Clinicians should therefore be alert to patient-reported changes in emotional responsiveness during SSRI or SNRI treatment, but the current evidence base cannot reliably quantify how often such changes reflect a direct pharmacological effect as opposed to residual symptoms of the underlying disorder.

Several findings bear on treatment decisions more broadly. High-certainty meta-analytic evidence confirms that antidepressants are superior to placebo for generalized anxiety disorder, with a number needed to treat of approximately seven, and that maintenance antidepressant therapy in stabilized OCD substantially reduces relapse risk. At the same time, fewer than half of antidepressant-treated patients in one observational cohort met conventional response thresholds for depression or anxiety within the first thirteen weeks, and distinct symptom trajectories suggest that early nonresponse — particularly in OCD — may warrant timely reassessment of the treatment plan rather than prolonged continuation of an ineffective regimen. The finding that dose reductions below 75% of the minimum effective dose were associated with clinically relevant withdrawal symptoms in roughly one-third of tapering intervals underscores the importance of gradual, carefully monitored discontinuation strategies.

Practitioners should also weigh the comparative tolerability profile of serotonergic antidepressants. One prospective study in drug-naïve patients with generalized anxiety disorder reported predominantly mild, self-limiting adverse reactions with no observed emotional blunting, while a secondary analysis of a randomized trial found that escitalopram produced significantly more adverse events than a mindfulness-based intervention despite comparable anxiolytic efficacy. Antidepressant-associated bruxism emerged in both a case report and a cross-sectional study, and high rates of early treatment abandonment — up to 34% within six months in one clinic sample — suggest that side-effect burden, whether emotional or somatic, may undermine adherence in routine practice.

Taken together, these observations suggest that clinicians should proactively inquire about subjective emotional changes during SSRI and SNRI treatment, consider validated instruments such as the Oxford Depression Questionnaire when emotional blunting is suspected, and integrate patient-reported emotional experience into shared decision-making about dose adjustment, augmentation, or switching. The absence of controlled trials specifically designed to distinguish medication-induced emotional blunting from illness-related affective restriction remains a critical gap; until such evidence is available, clinical judgments about the source of emotional dulling in any individual patient will necessarily remain provisional.

Effect Estimates by Study

Study	Measure	Value	95% CI	p-value
Sem Cohen, 2025	PPV	86	83–88	NR
Andrea Cipriani, 2026	adjusted relative risk	0.62	0.44–0.88	.007
Jeffrey A Mills, 2024	mean difference (standardized change in continuous measures of anxiety)	-1.98	-10.2–14.2	0.750
Tempei Otsubo, 2025	LS mean difference	-1.9	-3.4–-0.4	0.012
Gizem Nur Pala Avan, 2025	OR	5.928	3.708–9.478	< 0.001
Taro Kishi, 2025	RR	0.53	0.42–0.68	<0.01
Sem E Cohen, 2025	MD	2.65	1.85–3.46	< 0.0001
Prashant Babubhai Patel, 2026	partial η2	0.57	NR	< 0.001
Liang Xue, 2026	Cohen's d	0.73	NR	< 0.001
Hiroe Hu, 2024	MD	-1.6	-3.4–0.3	NR
Siegfried Kasper, 2024	adjusted mean change in MADRS total score	-12.1	-13.3–-11.0	< 0.01
Andri Rennwald, 2026	OR	3.2	1.4–7.4	NR
Katarina Kopcalic, 2025	RR	1.41	1.29–1.55	NR
Judith Cukor, 2024	RR	1.02	1.0–1.05	0.041
Hasan Mirzazadeh, 2025	Time X Group interaction F-statistic (Y-BOCS)	2.44	NR	0.12
Jurriaan M J L Brouwer, 2024	OR	1.34	0.89–2.02	not significant
Toshiaki Kikuchi, 2024	prevalence	67.1	65.5–68.7	NR
Luca Giacovelli, 2025	mean Y-BOCS total score change	NR	NR	< 0.01

Of the 30 included studies, 18 reported an extractable quantitative effect estimate and are listed above; the remaining 12 (case reports, study protocols, neuroimaging or qualitative analyses without a reported effect size) are omitted from this table and are flagged individually in Appendix B for reviewer follow-up.

3.3 Certainty of Evidence

Overall confidence in the available evidence is very low given the heterogeneity of study designs, indirectness relative to the prespecified PICO, imprecision, and risk of bias across the included studies.

4. Discussion

Summary of Evidence

This scoping review mapped the available evidence from 30 included articles on the relationship between SSRI or SNRI treatment and emotional blunting, reduced emotional authenticity, or subjective personality changes in adults with depressive or anxiety-spectrum disorders. The central finding is a pronounced mismatch between the clinical salience of the question and the volume of evidence that directly addresses it. Only two of the 30 included studies bore directly on the review's central question (one using a validated emotional-blunting instrument, one documenting treatment adherence patterns without specifically measuring blunting); the remaining 28 addressed adjacent topics — antidepressant efficacy, adverse-event profiling, neuroimaging correlates, pharmacogenomics, or treatment discontinuation — without measuring emotional blunting or related subjective emotional outcomes as a defined endpoint.

The single study that directly assessed emotional blunting prevalence, a cross-sectional web-based survey of over 3,300 Japanese adults with self-reported major depressive disorder, found that roughly two-thirds of antidepressant-treated respondents endorsed symptoms of emotional blunting, with severity correlating significantly with depression, anxiety, functional impairment, and quality of life. This estimate, however, rests on unconfirmed self-reported diagnoses and a cross-sectional design that cannot disentangle medication effects from residual illness. A second study documented high rates of early antidepressant abandonment without isolating emotional side effects as a driver.

The broader evidence set confirmed that SSRIs and SNRIs are efficacious for generalized anxiety disorder and obsessive-compulsive disorder, with high-certainty meta-analytic data supporting their use. Yet none of the efficacy-focused trials, meta-analyses, or neuroimaging studies employed validated emotional-blunting instruments or prespecified emotional blunting as an outcome. The absence of emotional blunting from most included studies likely reflects the fact that investigators were not looking for it rather than evidence that the phenomenon does not occur. This interpretive asymmetry — between a potentially common patient experience and a near-total absence of purpose-designed measurement — is the most striking pattern to emerge from the evidence map.

Limitations of this Review

Several methodological features of this review constrain the conclusions that can be drawn. The search was restricted to PMC Open Access articles, which may have introduced coverage bias by excluding relevant studies published in subscription-only journals, non-English-language sources, or grey literature repositories. This restriction means that foundational or landmark work on SSRI- and SNRI-associated emotional blunting published outside the open-access subset may not be represented.

A single reviewer conducted screening, data extraction, and quality assessment throughout, without independent dual screening or consensus arbitration. This approach increases the risk that eligible studies were missed at the screening stage or that extraction errors went undetected.

The heterogeneity of study designs, populations, outcome measures, and analytic frameworks across the 30 included articles precluded statistical pooling. The review therefore relies on narrative synthesis without meta-analysis, which limits the precision of any summary estimate and prevents formal assessment of between-study heterogeneity or publication bias. Given the scoping nature of this review, the intent was to map the breadth and character of the available evidence rather than to produce a pooled effect estimate, but readers should interpret the findings accordingly.

Implications for Future Research

The evidence gaps identified in this review point toward several concrete research priorities. No included study employed a controlled, prospective design with emotional blunting or reduced emotional responsiveness as a prespecified primary outcome. Randomized trials that measure emotional blunting longitudinally — using validated instruments such as the Oxford Depression Questionnaire — alongside standard efficacy and safety endpoints are needed to establish incidence, time course, and dose–response relationships.

Head-to-head comparisons of SSRIs versus SNRIs, and of serotonergic versus non-serotonergic antidepressants, with emotional blunting as a defined outcome would help clinicians make informed prescribing decisions. Trials incorporating a psychotherapy-only comparator arm are essential to distinguish medication-attributable emotional restriction from residual affective symptoms of the underlying disorder.

Population gaps are substantial. Adults with social anxiety disorder, panic disorder, and persistent depressive disorder were absent or poorly represented, as were older adults and patients with medical comorbidities. Future studies should recruit from these underrepresented groups. Outcomes specified in this review's eligibility criteria — including creativity, motivation, emotional empathy, relationship satisfaction, and antidepressant discontinuation attributable specifically to emotional side effects — were not examined in any included study and warrant dedicated investigation.

5. Conclusions

This scoping review identified 30 recent studies addressing SSRI or SNRI treatment in adults with depressive or anxiety-spectrum disorders, yet only one directly measured emotional blunting using a validated instrument, finding that a majority of antidepressant-treated adults with self-reported major depressive disorder endorsed such symptoms. The remaining evidence addressed treatment efficacy, safety, neurobiological correlates, or discontinuation patterns without specifically assessing changes in emotional authenticity or subjective personality. Available evidence therefore suggests that emotional blunting during serotonergic antidepressant use may be common, but the absence of any controlled prospective study designed with emotional blunting as a primary endpoint means the current literature cannot reliably distinguish medication-attributable emotional restriction from residual affective symptoms of the underlying disorder. Controlled trials employing validated emotional-blunting measures across diverse diagnostic populations and treatment durations are needed before firm clinical recommendations regarding this outcome can be made.

Methodological Limitations

1. Open-access full-text constraint: Full-text retrieval was restricted to freely accessible open-access sources and repositories. Paywalled subscription journals could not be retrieved through this review's framework. Where a reviewer-provided source was used to supplement retrieval for an individual article, that article is flagged in Appendix B for the reviewer's consideration before submission.

2. Single-reviewer relevance screening: Relevance screening was performed without a second independent reviewer. PRISMA 2020 expects dual-independent screening; the screening procedure used here applied a documented relevance threshold instead. Reviewers should review and override the included set as needed.

3. No backward citation chasing: This review did not include citation chasing (forward or backward) beyond the initial Boolean retrieval. Articles cited by included studies were not retrieved.

4. No grey literature: This review did not search grey literature sources (conference abstracts, dissertations, clinical-trial registry results-only entries, regulatory filings) beyond the structured ClinicalTrials.gov registry index.

5. Retrieval ceilings: Per-database retrieval is capped (PubMed: 10000; OpenAlex: 10000; ClinicalTrials.gov: 500). Queries returning more results than the ceiling will silently truncate to the cap.

6. Boolean query scope: The Boolean query is assembled as (Population) AND (Intervention/Exposure) only, per Cochrane §4.4.4. Comparator and Outcome filtering happens at the relevance-scoring stage, not at the search-query stage.

7. Publication year filter: Articles were restricted to those published between 2024 and 2026 (inclusive). Articles for which a publication year could not be confirmed were excluded from the analyzed set.

8. Retraction screening: PubMed and PMC metadata were inspected for retraction signals at retrieval time. Articles flagged as retracted are annotated in §3.1 Study Characteristics and surfaced for reviewer confirmation in Appendix B. Retracted articles are not automatically excluded — inclusion is a reviewer decision.

Protocol and Registration

This review was not registered in PROSPERO or another registry. A pre-specified protocol was generated alongside this manuscript and may be submitted post-hoc to PROSPERO or OSF for registration.

Funding

This review received no specific funding from any agency in the public, commercial, or not-for-profit sectors.

Conflicts of Interest

The authors declare no competing interests.

Data Availability

All extracted data and the verbatim Boolean query used for retrieval are available in the supplementary materials. Source articles are publicly accessible via PubMed / PubMed Central / open-access providers.

References

1. Sem Cohen,Jasper Brian Zantvoord,Ton de Boer,Taina Mattila,Guido van Wingen,Damiaan Denys Does early non-improvement predict treatment failure in pharmacotherapy for obsessive-compulsive disorder? A diagnostic test accuracy meta-analysis with individual participant data. Psychol Med. 2025;55:e237. doi:10.1017/S0033291725101335. PMID: 40814277.
2. Shalini Shukla,Naveen Kumar,Nikhil Goel,Navin Budania,Siddharth Ahuja,Vinod Kumar Bhardwaj ADVERSE DRUG REACTIONS OF SELECTIVE SEROTONIN REUPTAKE INHIBITORS IN GENERALIZED ANXIETY DISORDER: A PROSPECTIVE OBSERVATIONAL STUDY FROM A TERTIARY CARE HOSPITAL IN RURAL INDIA doi:10.53555/w0x2b258.
3. Andrea Cipriani,Karen Barros Parron Fernandes,Benoit H Mulsant,Orestis Efthimiou,Nicola Williams,Sam Mort,Rania Elgarf,Qiang Liu,Nyla Haque,Jennifer Potts,Roger Ede,Robin Fox,Marcos Liboni,Diego A Nesi Cavicchioli,Judit Simon,Katharine A Smith,Caroline Zangani,Zhenpeng Li,Ursula Taylor,M Ishrat Husain,Maddalena Cipriani,Paulo V Carpaneze Dalaqua,Eduardo Fernandes Leite,Gustavo Aliano Gâmbaro,Diogo Nabhan Silveira,Luísa Manfredin Vila,Fernanda Liboni Cavicchioli,Toshi A Furukawa,Huseyin Naci,Edoardo G Ostinelli A Decision-Support System to Personalize Antidepressant Treatment in Major Depressive Disorder: A Randomized Clinical Trial. JAMA. 2026;335(14):1219-1231. doi:10.1001/jama.2026.1327. PMID: 41779422.
4. Jeffrey A Mills,Eric Mendez,Jeffrey R Strawn The Impact of Development on Antidepressant and Placebo Response in Anxiety Disorders: A Bayesian Hierarchical Meta-Analytic Examination of Randomized Controlled Trials in Children, Adolescents, and Adults. J Child Adolesc Psychopharmacol. 2024;34(7):302-309. doi:10.1089/cap.2024.0016. PMID: 38800869.
5. Tempei Otsubo,Koichiro Watanabe,Takeshi Inoue,Ataru Inagaki,Toshiaki Kikuchi,Keisuke Nomoto,Shingo Higa,Yuko Asami,Bin Sun,Kristina Marschall A randomized, double-blinded, placebo-controlled study to evaluate the efficacy and safety of venlafaxine extended release in Japanese patients with generalized anxiety disorder. Psychiatry Clin Neurosci. 2025;79(12):849-858. doi:10.1111/pcn.13907. PMID: 41123332.
6. Dheeraj Rai,Doug Webb,Amanda Lewis,Leonora Cotton,Jade Eloise Norris,Regi Alexander,David S Baldwin,Traolach Brugha,Madeleine Cochrane,Maria Chiara Del Piccolo,Emma J Glasson,Katherine K Hatch,David Kessler,Peter E Langdon,Helen Leonard,Stephanie J MacNeill,Nicola Mills,Maximiliano Vazquez Morales,Zoe Morgan,Raja Mukherjee,Alba X Realpe,Ailsa Russell,Sergio Starkstein,Jodi Taylor,Nicholas Turner,Joanna Thorn,Jack Welch,Nicola Wiles Sertraline for anxiety in adults with a diagnosis of autism (STRATA): study protocol for a pragmatic, multicentre, double-blind, placebo-controlled randomised controlled trial. Trials. 2024;25(1):37. doi:10.1186/s13063-023-07847-3. PMID: 38212784.
7. Laura Orsolini,Silvia Bellagamba,Umberto Volpe Lurasidone as add-on to fluoxetine in obsessive-compulsive disorder with comorbid restrictive anorexia: a case report. Int Clin Psychopharmacol. 2024;39(3):211-214. doi:10.1097/YIC.0000000000000502. PMID: 37556307.
8. Anouk van der Straten,Willem Bruin,Laurens van de Mortel,Freek Ten Doesschate,Maarten J M Merkx,Pelle de Koning,Nienke Vulink,Martijn Figee,Odile A van den Heuvel,Damiaan Denys,Guido van Wingen Pharmacological and Psychological Treatment Have Common and Specific Effects on Brain Activity in Obsessive-Compulsive Disorder. Depress Anxiety. 2024;2024:6687657. doi:10.1155/2024/6687657. PMID: 40226646.
9. Gizem Nur Pala Avan,Ali Erdoğan,Buket Cinemre,Burak Kulaksızoğlu,Özmen Metin Bruxism and Sleep Disorders in Patients Diagnosed With Depressive Disorder and Anxiety Disorder Using Antidepressants. J Oral Rehabil. 2025;52(1):57-63. doi:10.1111/joor.13875. PMID: 39381861.
10. Yiding Han,Haohao Yan,Xiaoxiao Shan,Huabing Li,Feng Liu,Ping Li,Dongsheng Lv,Jingping Zhao,Wenbin Guo Decoding the striatum of drug-naive patients with obsessive-compulsive disorder: a transcriptome and longitudinal functional magnetic resonance imaging study. Transl Psychiatry. 2025;15(1):258. doi:10.1038/s41398-025-03475-4. PMID: 40721576.
11. Taro Kishi,Kenji Sakuma,Masakazu Hatano,Shun Hamanaka,Yasufumi Nishii,Nakao Iwata Relapse rates in stable obsessive-compulsive disorder after antidepressant discontinuation versus maintenance: A systematic review and meta-analysis. Psychol Med. 2025;55:e252. doi:10.1017/S0033291725101578. PMID: 40874319.
12. Sem E Cohen,Bram W Storosum,Jasper B Zantvoord,Taina K Mattila,Anthonius de Boer,Damiaan Denys Individual patient data meta-analysis of placebo-controlled trials of selective serotonin reuptake inhibitors submitted for regulatory approval in adult obsessive-compulsive disorder. Br J Psychiatry. 2025;227(4):680-687. doi:10.1192/bjp.2025.87. PMID: 40369939.
13. Prashant Babubhai Patel,Nisheet Patel,Rajesh Hadia,Aastha Dave,Yatri Dave,Niyati Jayeshbhai Patel Comparative Effectiveness of Pharmacotherapy, Cognitive Behavioural Therapy and their Combination for Depression and Anxiety in Newly Diagnosed Cancer Patients: A 6-week Prospective Observational Study Indian Journal of Palliative Care. 2026. doi:10.25259/ijpc_235_2025.
14. Komal Kohat,Santenna Chenchula Sertraline-induced bruxism: a case report and review of the literature. J Med Case Rep. 2026;20(1). doi:10.1186/s13256-026-05834-1. PMID: 41673731.
15. Liang Xue,Hongfen Ni,Shuqi Xu Baseline serum oxytocin and OXTR rs53576 genotype are not predictive of escitalopram response in generalized anxiety disorder. Sci Rep. 2026;16(1). doi:10.1038/s41598-026-36366-9. PMID: 41644973.
16. Hiroe Hu,Mihriye Mete,Neil K Rustgi,Charisma I Washington,Kavya Sanghavi,Mary Ann Dutton,Naomi M Simon,Amanda W Baker,Eric Bui,Elizabeth A Hoge Mindfulness Meditation vs Escitalopram for Treatment of Anxiety Disorders: Secondary Analysis of a Randomized Clinical Trial. JAMA Netw Open. 2024;7(10):e2438453. doi:10.1001/jamanetworkopen.2024.38453. PMID: 39382900.
17. Rachael Frost,Sukvinder Kaur Bhamra,Michael Heinrich Herbal Products and Antidepressants: A Safe Combination or a Risky Mix? Phytotherapy Research. 2026. doi:10.1002/ptr.70173.
18. Siegfried Kasper,Erich Seifritz,Hans‐Peter Volz Efficacy of Silexan in Patients with a Major Depressive Episode – First Results from a Multi-centre, Double-blind, Randomised, Placebo- and Reference-controlled Phase III Trial European Psychiatry. 2024. doi:10.1192/j.eurpsy.2024.479.
19. Andri Rennwald,Mark A Horowitz,Oliver Senn,Oliver Neuner-Jehle,Michael P. Hengartner Incidence of antidepressant withdrawal reactions: A prospective longitudinal cohort study in primary care patients medRxiv. 2026. doi:10.64898/2026.05.12.26352975.
20. Katarina Kopcalic,Justin Arcaro,Antonio Pinto,Shehzad Ali,Corrado Barbui,Chiara Curatoli,Janet Martin,Giuseppe Guaiana Antidepressants versus placebo for generalised anxiety disorder (GAD). Cochrane Database Syst Rev. 2025;1(1):CD012942. doi:10.1002/14651858.CD012942.pub2. PMID: 39880377.
21. Judith Cukor,Zhenxing Xu,Veer Vekaria,Fei Wang,Mark Olfson,Samprit Banerjee,Gregory E. Simon,George S. Alexopoulos,Jyotishman Pathak Longitudinal trajectories of symptom change during antidepressant treatment among managed care patients with depression and anxiety npj Mental Health Research. 2024. doi:10.1038/s44184-024-00104-8.
22. T. Jiménez Aparicio,M. Fernández Lozano,Milton Merizalde Torres,A. Rodríguez Campos Evaluation of adherence to treatment in patients with anxious-depressive syndrome. European Psychiatry. 2024. doi:10.1192/j.eurpsy.2024.1403.
23. Hasan Mirzazadeh,Younes Ghaeminia,Amin Mohamad Niaei,Mohammadsadegh Kamran,Leila Razeghian Jahromi,Reza Moshfeghinia Memantine augmentation of escitalopram in treatment of executive function among patients with obsessive-compulsive disorder (OCD): a double-blind placebo-controlled randomized clinical trial. BMC Psychiatry. 2025;25(1):561. doi:10.1186/s12888-025-06856-7. PMID: 40457233.
24. Francesco Monaco,Annarita Vignapiano,Ernesta Panarello,Stefania Landi,Giuseppe Scarano,Giovanna Celia,Giulio Corrivetti,Luca Steardo,Mauro Cozzolino Lurasidone and Fluvoxamine Combination in Eating Disorders with Comorbid Obsessive-Compulsive Disorder: Preliminary Evidence from an Observational Study. Med Sci (Basel). 2025;14(1). doi:10.3390/medsci14010008. PMID: 41562898.
25. Jurriaan M J L Brouwer,Klaas J Wardenaar,Ilja M Nolte,Edith J Liemburg,Pierre M Bet,Harold Snieder,Hans Mulder,Danielle C Cath,Brenda W J H Penninx Association of CYP2D6 and CYP2C19 metabolizer status with switching and discontinuing antidepressant drugs: an exploratory study. BMC Psychiatry. 2024;24(1):394. doi:10.1186/s12888-024-05764-6. PMID: 38797832.
26. Toshiaki Kikuchi,Jun-Ichi Iga,Masato Oosawa,Tatsuya Hoshino,Yoshiya Moriguchi,Miwa Izutsu A web-based survey on the occurrence of emotional blunting in patients with major depressive disorder in Japan: Patient perceptions and attitudes. Neuropsychopharmacol Rep. 2024;44(2):321-332. doi:10.1002/npr2.12417. PMID: 38616339.
27. Sucharita Mandal,Kaushik Mukhopadhyay,Aparajita Guin,Rupal Sorkhel,Pulama Mitra,Bisweswar Ojha,Kamini Verma,Sukriti Mukherjee Efficacy and safety profile of vortioxetine as an add-on molecule in obsessive-compulsive disorder: a randomized double-blinded placebo-controlled superiority trial - a study protocol. BMC Psychiatry. 2025;25(1):1116. doi:10.1186/s12888-025-07374-2. PMID: 41286906.
28. Amanda Hazel Dilmore,Rayus Kuplicki,Daniel McDonald,Megha Kumar,Mehrbod Estaki,Nicholas Youngblut,Alexander Tyakht,Gail Ackermann,Colette Blach,Siamak MahmoudianDehkordi,Boadie W Dunlop,Sudeepa Bhattacharyya,Salvador Guinjoan,Pooja Mandaviya,Ruth E Ley,Rima Kaddaruh-Dauok,Martin P Paulus,Rob Knight Medication use is associated with distinct microbial features in anxiety and depression. Mol Psychiatry. 2025;30(6):2545-2557. doi:10.1038/s41380-024-02857-2. PMID: 39794490.
29. Luca Giacovelli,Eleonora Piccoli,Paola Landi,Matteo Vismara,Beatrice Benatti,Bernardo Dell'Osso Brexpiprazole augmentation in treatment-resistant obsessive-compulsive disorder: a preliminary retrospective observational study. Int Clin Psychopharmacol. 2025;40(6):e14-e20. doi:10.1097/YIC.0000000000000583. PMID: 39908230.
30. Diana Dubrall,P Christ,Matthias Schmid,Bernhardt Sachs,Catharina Scholl An analysis of spontaneous adverse drug reaction reports of psychiatric patients stratified by reporter type provides differentiated data European Psychiatry. 2025. doi:10.1192/j.eurpsy.2025.553.

Appendix A — PRISMA 2020 Checklist

The PRISMA 2020 flow of records (identification, screening, eligibility, and inclusion) and the checklist coverage for this review are summarized in the PRISMA 2020 Flow Diagram and the Coverage and Validity Scorecard above, and throughout the Methods and Results sections.

Appendix B — Reviewer Worklist

The following items, noted throughout the review, are collected here for reviewer follow-up before final submission:

• Studies without an extractable quantitative effect estimate (12 of 30): case reports, study protocols without results, and neuroimaging or qualitative analyses that did not report an effect size are omitted from the §3 Effect Estimates by Study table and are flagged here for reviewer confirmation.
• Reviewer-/author-provided sources: where a reviewer-provided source supplemented open-access retrieval for an individual article (Author's reference collection), that article is flagged for the reviewer's consideration before submission (see Pipeline Limitations 1).
• Single-reviewer screening: relevance screening, data extraction, and quality assessment were performed by a single reviewer without second-reviewer arbitration; the reviewer should review and override the included set as needed (see Pipeline Limitations 2).
• Retraction screening: articles flagged for retraction signals at retrieval are not autonomously excluded; inclusion of any retraction-flagged article remains a reviewer decision (see Pipeline Limitations 8).

Suggested citation

Krasnova M. Evidence Review: Emotional blunting and subjective personality change during SSRI and SNRI treatment in adults with depression or anxiety margaritakrasnovamd.com. Last reviewed 2026-05-28. Available at: https://margaritakrasnovamd.com/evidence-review/antidepressants-emotional-blunting.html