Psychiatric and Sociodemographic Associations with Bariatric Surgery Outcomes in a Mexican Tertiary Care Center

Abstract

Background: Obesity is highly prevalent in Mexico and frequently coexists with psychiatric disorders that may influence postoperative outcomes in bariatric surgery candidates.

Materials and methods: We conducted an observational, descriptive, retrospective study including 123 adult patients evaluated by the Psychiatry Service as part of an institutional bariatric surgery protocol at the National Medical Center “20 de Noviembre” between 2020 and 2024. Of these, 117 (95.1%) underwent the surgical procedure; postoperative weight loss and adherence analyses were restricted to operated patients. Psychiatric diagnoses were coded according to ICD-10. Descriptive and inferential statistics were applied; all comparisons were unadjusted.

Results: Psychiatric diagnoses were present in 31.7% of patients. Those with a psychiatric disorder showed significantly lower absolute postoperative weight loss than those without (10.87 kg vs. 20.97 kg; p < 0.001; Cohen's d = 0.80) and a higher prevalence of low adherence (p = 0.047). More than half of patients with a psychiatric history experienced documented postoperative symptom improvement. Given the heterogeneous follow-up duration, the weight loss figures should be interpreted as exploratory.

Conclusion: Psychiatric disorders are common among bariatric surgery candidates at this institution and are associated with reduced postoperative weight loss and lower adherence in unadjusted analyses. Structured psychiatric evaluation should be integrated into bariatric care pathways; confirmatory studies with multivariable adjustment and standardized follow-up periods are needed.

Graphical abstract

Keywords

bariatric surgery / mental disorders / obesity / medication adherence / psychiatric comorbidity / Mexico

Introduction

Obesity is a chronic, multifactorial disease and a leading global public health burden owing to its associations with metabolic, cardiovascular, and psychiatric disorders[1]. Bariatric surgery is the most effective intervention for achieving sustained weight reduction and improving metabolic comorbidities in individuals with severe obesity[2,3]. Despite well-documented somatic benefits, the interaction between bariatric surgery and mental health remains an area of active investigation and ongoing controversy[4].

Up to 55% of bariatric surgery candidates present at least one preoperative psychiatric disorder, most commonly depression and anxiety[4]. While many patients experience postoperative improvements in psychiatric symptoms, others show increased rates of affective disorders, disordered eating behaviors , and heightened suicide risk[5,6]. Proposed contributors include rapid weight loss, body-image changes, and neurobiological shifts in gut-brain signalling; nutritional deficiencies—particularly of vitamin B12 and vitamin D—have also been associated with postoperative mood and cognitive symptoms[7].

In Mexico, approximately 33.3% of adults have obesity, placing the country among those with the highest prevalence worldwide[8]. Bariatric procedures have increased substantially in recent years[2], yet evidence on psychiatric comorbidities in this population remains limited[8]. A bidirectional relationship between obesity and psychiatric disorders is well established, with binge eating disorder, major depressive disorder, anxiety disorders, and personality disorders being the most prevalent psychiatric conditions among surgical candidates[4,9]. Preoperative psychiatric disorders can negatively affect adherence to postoperative lifestyle modifications and are associated with attenuated weight-loss outcomes[6,9].

Research on postoperative mental-health trajectories reports mixed findings: meta-analyses document improvements in depressive and anxiety symptoms during the first postoperative year in approximately 50% of patients[6], whereas other studies identify increased prevalence of psychiatric disorders and heightened suicide risk in longer follow-up periods[4,10]. Identifying sociodemographic and psychiatric factors associated with postoperative outcomes may strengthen preoperative evaluation protocols and inform targeted follow-up strategies, particularly in underserved populations.

The primary aim of this study was to describe the sociodemographic and clinical profile of patients evaluated by the Psychiatry Service as part of an institutional bariatric surgery protocol. Specific objectives were to: (1) characterize key sociodemographic variables; (2) determine the prevalence of preoperative psychiatric diagnoses; (3) examine associations between sociodemographic variables and psychiatric diagnoses; (4) explore whether psychiatric prevalence differed by surgery type; and (5) describe the associations of preoperative psychiatric diagnosis with postoperative weight loss and adherence to follow-up recommendations among patients who underwent surgery. All analyses were unadjusted and exploratory; the study was not designed to identify independent predictors of bariatric outcomes.

Materials and methods

Study design and setting

This was an observational, descriptive, retrospective cohort study conducted at the National Medical Center “20 de Noviembre” (CMN 20 de Noviembre), ISSSTE, Mexico City. Medical records of patients evaluated by the Psychiatry Service as part of the interdisciplinary bariatric surgery protocol were reviewed.

Participants

Eligible patients were adults (≥ 18 years) who (1) were evaluated by the Psychiatry Service between January 2020 and December 2024; (2) met criteria for obesity grade II or III (body mass index (BMI) ≥ 35 kg/m² with comorbidities or ≥ 40 kg/m²)[11]; (3) had complete medical records with sociodemographic and psychiatric information; and (4) had at least one documented postoperative follow-up visit if they underwent surgery. Exclusion criteria were: incomplete records lacking essential psychiatric or postoperative data; surgery performed outside the institution; procedures indicated for non-obesity purposes; severe uncontrolled psychiatric disorders contraindicating surgery (e.g., active schizophrenia and acute bipolar episode); revisional procedures; and follow-up periods shorter than the observation requirement. Cases with inconsistent or insufficient postoperative information were eliminated.

Sample size

Theoretical sample size estimation, based on an expected 50% prevalence of psychiatric disorders, 95% confidence interval, and 5% precision, indicated a requirement of 384 patients for an infinite population; finite-population adjustment for an estimated 500 eligible patients yielded a target of 217. Due to the retrospective design and available records, all 123 patients meeting inclusion criteria were analyzed, representing approximately 57% of the target sample. This underpowering limits statistical power for subgroup comparisons with small cell sizes and should be considered when interpreting results.

Procedures and variables

Psychiatric diagnoses were assigned using the International Classification of Diseases, 10th Revision (ICD-10)[12]. More than one ICD-10 diagnosis could be assigned per patient; frequencies therefore reflected individual diagnoses and were not mutually exclusive. Variables collected included sociodemographic data (age, sex, marital status, educational level, occupation, socioeconomic status, and state of residence), clinical data (BMI, duration of obesity, comorbidities and type of bariatric procedure), and psychiatric data (preoperative diagnosis, ICD-10 code, number of consultations, treatment type, substance use, postoperative symptom changes, and treatment adherence). Postoperative improvement was assessed from clinical documentation rather than standardized validated scales; accordingly the term “documented improvement” is used throughout.

Postoperative outcomes

Of the 123 patients included, 117 (95.1%) underwent the surgical procedure; 6 (4.9%) had not been operated on at the time of data collection. Postoperative weight-loss and adherence analyses were performed exclusively in the 117 operated patients. Weight loss is expressed as absolute kilograms (kg) because the variability in follow-up duration across patients (range not standardised) precluded reporting percentage total weight loss (%TWL) or percentage excess weight loss (%EWL); these figures should therefore be interpreted as exploratory rather than definitive outcome measures. Missing postoperative data were handled by complete-case analysis. Adherence was categorized as good, moderate, or low based on clinical notes, rather than on objective metrics.

Statistical analysis

Descriptive statistics included means ± standard deviations (SD) and ranges for quantitative variables, and absolute frequencies with percentages (with 95% confidence intervals [CI] where applicable) for categorical variables. Normality was assessed with Kolmogorov–Smirnov and Shapiro–Wilk tests. Non-parametric quantitative comparisons used Mann–Whitney U and Kruskal–Wallis tests; categorical associations used Chi-square or Fisher’s exact tests. Effect sizes were calculated as Cohen's d (continuous outcomes) and r (rank-based). A two-tailed p < 0.05 was considered statistically significant. All analyses were unadjusted and exploratory; no multivariable regression models were fitted. Results are therefore interpreted as descriptive associations rather than independent predictors. Analyses were performed with SPSS version 25.0 (IBM Corp., Armonk, NY).

Results

Sociodemographic and clinical profile

A total of 123 patients evaluated by the Psychiatry Service were analyzed (Table 1). Of these, 117 (95.1%) underwent the surgical procedure; 6 (4.9%) had not been operated on at data cut-off. The sample had a mean age of 45.2 years (SD 9.35; range 20–67). Females predominated (n = 93, 75.6%; 95% CI: 67.3–82.4), with males accounting for 24.4% (n = 30). The mean preoperative BMI was 49.1 kg/m² (SD 7.69; range 32.8–71.0), and mean duration of obesity was 23.7 years (SD 12.48; range 2–59).

Most patients had higher education (78.0%), followed by middle (17.1%) and basic (4.9%) levels. Regarding occupation, 39.0% were formally employed, 23.6% informally, and 37.4% were unemployed or homemakers. Socioeconomic status was middle in 50.0%, low in 33.6%, and high in 17.1%. The most represented states of origin were Mexico City (26.8%) and Sinaloa (18.7%), reflecting the national referral coverage of the hospital.

The most frequent surgical technique was vertical sleeve gastrectomy (n = 66; 53.7%), followed by gastric bypass (n = 31; 25.2%). The most common metabolic comorbidities were hypertension (48.8%; 95% CI: 40.1–57.5), type 2 diabetes (35.8%; 95% CI: 27.9–44.6), and dyslipidaemia (30.9%; 95% CI: 23.4–39.5). Non-metabolic conditions included fatty liver disease (30.9%), insulin resistance (26.0%), hypothyroidism (24.4%), and chronic/erosive gastritis (17.9%).

Preoperative psychiatric profile

Preoperative psychiatric diagnoses (ICD-10) were absent in 68.3% of patients (n = 84); 31.7% (n = 39) had at least one diagnosis (Table 2). The most common diagnoses were generalised anxiety and other anxiety disorders (F41; 12.2%) and depressive episode (F32; 12.2%), followed by stress-related disorders (F43; 6.5%), eating disorders (F50; 4.1%), persistent affective disorders (F34; 1.6%), and isolated cases of bipolar disorder (F31), obsessive-compulsive disorder (F42), and recurrent depressive disorder (F33), each 0.8%. Multiple diagnoses per patient were permitted; frequencies are therefore not mutually exclusive.

Topiramate was prescribed to 89.4% of patients (generally 100 mg/day). Antidepressants were used by 29.3%, most commonly fluoxetine (9.8%), sertraline (8.1%), and escitalopram (6.5%). Non-pharmacological interventions included psychoeducation, brief psychotherapy, or emotional validation. Alcohol use was reported by 47.2% (social or occasional in 34.2%; active in 13.0%); tobacco use in 37.4%; other substances in 3.3%.

Postoperative psychiatric evolution

Among the 117 operated patients, 67 (54.5%) experienced documented postoperative improvement, 32 (26.0%) showed no notable change, and 9 (7.3%) experienced worsening or development of new psychiatric symptoms. In 15 patients (12.2%), medical records did not allow conclusive evaluation due to ambiguous documentation or insufficient follow-up. Given that assessment relied on clinical notes rather than standardised instruments, these figures should be interpreted with caution. Improvements were most frequently noted in binge or emotional eating (33.3%), anxiety or depressive symptoms (30.9%), impulse control (23.6%), and overall mood stabilisation (20.3%). Persistent or worsened symptoms included food-related anxiety (9.8%), relapse of disordered eating or weight regain (5.7%), and medication-related adverse effects such as topiramate-associated irritability (4.1%).

Adherence

Among the 123 patients evaluated, good adherence was recorded in 95 (77.2%; 95% CI: 69.1–83.8), moderate adherence in 17 (13.8%; 95% CI: 8.8–21.0), and low adherence in 11 (8.9%; 95% CI: 5.1–15.3).

Association between psychiatric diagnosis and postoperative outcomes

Among the 117 operated patients, those with any preoperative psychiatric diagnosis showed lower absolute postoperative weight loss than those without a diagnosis (mean 10.87 kg vs. 20.97 kg; mean difference 10.10 kg; p < 0.001; Cohen's d = 0.80) (Figure 1). Given the heterogeneous follow-up duration across patients, these figures represent exploratory estimates. Although individual subgroup analyses were limited by small cell sizes, the same directional trend was observed across diagnostic categories, reaching statistical significance only for depressive episode (F32) and eating disorders (F50) in post-hoc comparisons. In exploratory subgroup analysis, mean postoperative weight loss was 10.9 kg (SD 10.2; n = 15) for depressive disorders (F32; Bonferroni-corrected p = 0.029) and 7.2 kg (SD 6.4; n = 5) for eating disorders (F50; corrected p = 0.032), compared with 21.0 kg (SD 14.3; n = 78) in patients without a psychiatric diagnosis; anxiety disorders (F41; 10.8 kg, n = 15) and stress-related disorders (F43; 11.4 kg, n = 8) were numerically lower but did not reach significance after correction. Because individual diagnostic subgroups were small, these subgroup estimates are exploratory.

Analysis of adherence and weight loss using the Kruskal–Wallis test revealed statistically significant differences across adherence groups (H = 9.54, p = 0.009): mean weight loss was 19.28 kg (SD 13.78) in the good-adherence group, 13.00 kg (SD 10.95) in the moderate group, and 10.05 kg (SD 8.77) in the low-adherence group. A statistically significant association was found between psychiatric diagnosis and adherence category (χ² = 6.11, p = 0.047): 27.4% of patients with good adherence carried a psychiatric diagnosis, compared with 35.3% in the moderate group and 63.6% in the low-adherence group (Figure 2). No significant differences were found between specific ICD-10 diagnoses and adherence categories (χ² = 14.45, p = 0.417).

Discussion

Patients with a preoperative psychiatric diagnosis had a mean postoperative absolute weight loss that was 10.10 kg lower than that of patients without a diagnosis (10.87 kg vs. 20.97 kg; p < 0.001; d = 0.80). Although these comparisons are unadjusted and the follow-up duration was heterogeneous across patients, the direction and magnitude of the association are consistent with prior evidence documenting attenuated weight-loss trajectories in bariatric candidates with affective and eating disorders[9].

The female predominance (75.6%) and mean age of 45.2 years are consistent with published demographic profiles of bariatric surgery cohorts[2,13]. The concentration of patients with higher education (78.0%) and formal employment (39.0%) likely reflects the profile of individuals accessing a high-speciality public institution within the Mexican Social Security system. Education and socioeconomic status have been described as modulators of adherence and postoperative outcomes[4].

A preoperative psychiatric diagnosis was present in 31.7% of patients, a rate lower than the 50%–70% reported internationally[4,5]. This discrepancy may reflect under-documentation in clinical records, evaluation constraints inherent to the routine protocol, or sociocultural barriers to psychiatric disclosure. Even so, the observed psychiatric prevalence was associated with significant differences in postoperative weight loss and adherence.

Post-hoc subgroup analyses identified statistically significant associations between lower weight loss and F32 (depressive episode) and F50 (eating disorders), consistent with evidence that major depression and binge-eating disorder are associated with poorer weight-loss outcomes and higher risk of weight regain[9,14]. Barbuti et al. further demonstrated that emotional dysregulation and lifetime binge eating disorder predicted insufficient weight loss at one-year follow-up after bariatric surgery[9]. It should be emphasised, however, that the analyses in the present study were unadjusted and did not control for age, sex, baseline BMI, surgery type, follow-up duration, or psychosocial support; causal inferences cannot therefore be drawn.

More than half of patients with a psychiatric history (54.5%) experienced documented postoperative improvement, particularly in binge eating, anxiety, and mood. This mirrors prior meta-analytic data reporting depressive and anxious symptom reductions in approximately 50% of patients during the first postoperative year[6,15]. Nonetheless, 7.3% experienced symptom worsening or new psychiatric diagnoses, reinforcing the importance of longitudinal follow-up[4]. Because assessments relied on clinical documentation rather than validated instruments, these findings should be interpreted as preliminary; the term “documented improvement” is used deliberately throughout.

The predominance of topiramate (89.4%) is notable and reflects the specific interdisciplinary protocol at this institution, aimed at appetite reduction, impulse control, and mood stabilisation in a population with high rates of binge eating and emotional dysregulation[16]. Randomized evidence supports topiramate for binge-eating disorder associated with obesity[16], although prescribing rates of this magnitude are uncommon in comparable cohorts.

A statistically significant association was found between psychiatric diagnosis and lower postoperative adherence (χ² = 6.11, p = 0.047), consistent with literature indicating that depressive symptoms and emotional dysregulation negatively affect therapeutic adherence[17]. Adherence may partly mediate the association between psychiatric diagnosis and lower weight loss—a pathway that warrants formal mediation analysis in future studies. No significant differences were found between specific ICD-10 diagnoses and adherence categories, suggesting that transdiagnostic dimensions such as motivation, impulsivity, and social support may be more relevant than categorical diagnosis alone.

Good adherence (77.2%) was associated with greater weight loss (19.28 kg), consistent with evidence that postoperative follow-up participation significantly improves bariatric outcomes[18,13]. The low prevalence of other substance use (3.3%) contrasts with international reports of increased post-surgical substance use[6,15]; this may reflect a genuinely low-risk population, underreporting, or insufficient follow-up time to capture such changes.

A recent analysis from the American Society for Metabolic and Bariatric Surgery 2025 Annual Meeting compared bariatric surgery with GLP-1 receptor agonist therapy and found that surgical patients had a lower five-year risk of anxiety, cognitive impairment, and substance-use disorders[19], reinforcing the dual benefit of surgery on weight and mental health outcomes. These findings contextualize the importance of integrating psychiatric evaluation within bariatric protocols.

Limitations

Several limitations must be considered. First, the retrospective design relies on the completeness of medical records, which may have led to under-reporting of psychiatric diagnoses, substance use, or symptom evolution. Second, the study was conducted at a single high-speciality public hospital, limiting generalisability to other settings. Third, psychiatric assessments were based on routine clinical evaluations rather than standardised diagnostic interviews, which may have introduced variability in diagnostic accuracy. Fourth, postoperative follow-up duration was heterogeneous across patients; absolute weight loss in kilograms therefore represents an imperfect and exploratory outcome measure, and reporting of %TWL or %EWL was not feasible. Fifth, adherence was categorised using clinical-note judgement rather than objective metrics. Sixth, the achieved sample (n = 123) represents approximately 57% of the estimated required sample (n = 217), limiting statistical power for subgroup comparisons. Seventh, all analyses were unadjusted; confounders including follow-up duration, surgery type, psychosocial support, and baseline metabolic status were not controlled for, and causal conclusions cannot be drawn.

Conclusion

This study describes the sociodemographic, clinical, and psychiatric profile of patients evaluated for bariatric surgery at a Mexican high-speciality centre. In unadjusted analyses, psychiatric disorders were associated with lower postoperative weight loss and reduced adherence. These findings are preliminary and require confirmation in studies with larger samples, standardised follow-up periods, and multivariable adjustment for relevant confounders. Structured preoperative psychiatric assessment should be a standard component of bariatric protocols, and long-term multidisciplinary follow-up integrating surgical, nutritional, and mental health teams is essential for optimizing sustained outcomes.

References

[1] Müller A, Mitchell JE, Sondag C, de Zwaan M. Psychiatric aspects of bariatric surgery. Curr Psychiatry Rep. 2013;15(10):397.

[2] Maldonado Pintado DG, Márquez González SM, Wimber Arellano M, Herrera MF. Cirugía bariátrica en méxico: características de la práctica en 2019. Cir Gen. 2022;44:134-142.

[3] O’Brien PE, Dixon JB, Brown W, et al. The laparoscopic adjustable gastric band (lap-band): a prospective study of medium-term effects on weight, health and quality of life. Obes Surg. 2002;12(5):652-660.

[4] Becerra AZ, Khalid SI, Morgenstern AS, et al. The association between bariatric surgery and psychiatric disorders: a national cohort study. Obes Surg. 2022;32(4):1110-1118.

[5] Olguín VP, Carvajal AD, Fuentes SM. Patología psiquiátrica y cirugía bariátrica. Rev Chil Cir. 2015;67(4):441-447.

[6] Castaneda D, Popov VB, Wander P, Thompson CC. Risk of suicide and self-harm is increased after bariatric surgery-a systematic review and meta-analysis. Obes Surg. 2019;29(1):322-333.

[7] Xanthakos SA. Nutritional deficiencies in obesity and after bariatric surgery. Pediatr Clin North Am. 2009;56(5):1105-1121.

[8] Guilbert L, Joo P, Ortiz C, et al. Safety and efficacy of bariatric surgery in mexico: a detailed analysis of 500 surgeries performed at a high-volume center. Rev Gastroenterol Mex (Engl Ed). 2019;84(3):296-302.

[9] Barbuti M, Carignani G, Weiss F, et al. Eating disorders and emotional dysregulation are associated with insufficient weight loss after bariatric surgery: a 1-year observational follow-up study. Eat Weight Disord. 2023;28(1):49.

[10] de Zwaan M, Enderle J, Wagner S, et al. Anxiety and depression in bariatric surgery patients: a prospective, follow-up study using structured clinical interviews. J Affect Disord. 2011 Sep;133(1-2):61-68.

[11] World Health Organization. Obesity: preventing and managing the global epidemic. report of a WHO consultation. World Health Organ Tech Rep Ser. 2000;894:i-xii, 1-253.

[12] World Health Organization. International classification of diseases, 10th revision (ICD-10). Geneva: WHO; 1992. ISBN:92-4-154419-8.

[13] Lima RC, Rodrigues TMDS, Scheibe CL, et al. Weight loss and adherence to postoperative follow-up after vertical gastrectomy for obesity treatment. Acta Cir Bras. 2021 Feb 22;36(2):e360203.

[14] Mauro MFFP, Papelbaum M, Brasil MAA, et al. Is weight regain after bariatric surgery associated with psychiatric comorbidity? a systematic review and meta-analysis. Obes Rev. 2019 Oct;20(10):1413-1425.

[15] Vermeer KJ, Monpellier VM, Cahn W, Janssen IMC. Bariatric surgery in patients with psychiatric comorbidity: significant weight loss and improvement of physical quality of life. Clin Obes. 2020 Aug;10(4):e12373.

[16] McElroy SL, Hudson JI, Capece JA, et al. Topiramate for the treatment of binge eating disorder associated with obesity: a placebo-controlled study. Biol Psychiatry. 2007 May 1;61(9):1039-1048.

[17] da Silva SS, Maia Âda C. Patients’ experiences after bariatric surgery: a qualitative study at 12-month follow-up. Clin Obes. 2013 Dec;3(6):185-193.

[18] Livhits M, Mercado C, Yermilov I, et al. Behavioral factors associated with successful weight loss after gastric bypass. Am Surg. 2010 Oct;76(10):1139-1142.

[19] Attia A, Toraih E, Baker J, Hennings D, Levy S. Psychiatric disorders less likely after weight-loss surgery than treatment with GLP-1s [abstract]. Presented at: American Society for Metabolic and Bariatric Surgery (ASMBS) Annual Meeting; 2025.

Rights & permissions

The Author(s) 2026.