Gender disparities in hospital admission patterns during the COVID-19 health emergency: a systematic review

Abstract

Background

Sex/gender differences influence health outcomes, healthcare access, and hospital use. The COVID-19 pandemic disrupted health systems and may have exacerbated existing disparities. This systematic review was conducted to examine whether sex/gender disparities exist across diagnostic categories and hospital admission routes, and to determine whether these differences were altered following the declaration of the COVID-19 pandemic.

Methods

This review was conducted according to the PRISMA guidelines. Searches were conducted in PubMed, Web of Science and Cochrane Library to identify studies published in English or Spanish between 2020 and 2024 examining hospital admissions before and during the COVID-19 pandemic, with sex-disaggregated data.

Results

A total of 41 studies met the inclusion criteria, revealing gender-related differences in hospital admissions during the pandemic. The articles were classified according to ICD-10 chapters. During the pandemic, among adults and across all age groups, there was a notable increase in hospitalisations among women for acute burns, alcohol-associated hepatitis, resected lung cancer, malignant melanoma, and others. Women also showed increased emergency visits for infections, mental health problems, and injuries. In contrast, men experienced an increase in admissions for gastrointestinal bleeding. Additionally, studies reported rises in sexual abuse of girls, higher self-harm rates among boys, and more admissions for mental health problems among girls.

Conclusions

This systematic review identified differences in hospital admissions for various conditions and highlighted social and health inequalities exacerbated by lockdown. These findings undersore the importance of integrating a gender perspective into public health strategies and responses during health emergencies.

Trial registration

The review protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO) on 22 August 2025 (CRD420251038282).

Background

According to the World Health Organization (WHO), males and females differ in biology, societal roles, and positions within family and social structures. These differences influence their health-related behaviors and how the healthcare system responds to their needs [1]. Although sex and gender are often examined separately, their interdependence shapes individual identity, cognitive function, and disease patterns. Some scholars advocate for the combined term “sex/gender” to reflect this interaction more accurately [2].

Access to healthcare services and patterns of use also vary by gender [3]. Disparities in healthcare access are not always explained by differences in health status [4]. Evidence shows that females often encounter more barriers than males in accessing effective care [5, 6]. Planned hospital admissions are typically preventive and signal effective health management, whereas unplanned admissions often reveal systemic failures in early disease detection [7]. In Spain, in 2022, females aged 15–79 years used primary care emergency departments more frequently than males [8]. These findings support the need for a gender-informed approach to improve care effectiveness and equity.

The COVID-19 pandemic created unprecedented disruptions in global healthcare systems [9]. Emergency measures altered patients’ care-seeking behaviors, while health services restructured operations, postponed elective procedures, and reallocated resources [9–11]. The consequences of such restructuration could increase previous gender inequalities in hospital access. For instance, non-COVID excess mortality was three times higher in females than in males during the pandemic’s first wave in Spain, suggesting diagnostic and treatment delays or inadequate management of chronic diseases [12]. A related study in Italy, using hospital registry data, found that hospital utilization declined during early 2020 relative to 2018–2019, particularly among females [13].

In this sense, epidemiology has been widely recognized as a key instrument for identifying social inequities in health, by allowing the detection of systematic differences derived from structural, economic and cultural factors [14]. Among these axes of inequality, gender constitutes a central dimension, since it conditions the patterns of exposure, vulnerability and use of health services [15]. Peters et al. (2023) advocate not only to disaggregate data by sex, but also to use one sex as a control group for the other. The aim of that approach is to identify disparities, understand how sex/gender interacts with other variables, and allocate resources more equitably to distinguish between biological or social mechanisms that place women or men in contexts of health disadvantage [16]. With this in mind, various records such as the admission rate, length of stay, mortality and reason for hospitalization according to the International Classification of Diseases (ICD) are usually studied worldwide, but they are rarely exploited from a gender perspective. Recent studies have directly compared in-hospital admission and mortality rates between women and men to identify diseases characterized by gender-based inequalities, and to propose modifiable key factors contributing to these disparities [17, 18].

Building on these findings, it is essential to examine the evolution of hospital admission ratios for various diseases among female and male patients from the pre-pandemic period through the COVID-19 pandemic, in order to elucidate the impact of the pandemic health emergency on gender disparities. Accordingly, we aim to systematically review the presence of sex/gender disparities across diagnostic categories and hospital outcomes, and to determine whether these differences changed following the onset of the COVID-19 pandemic.

Methods

Aim

This review aims to examine whether sex/gender disparities exist across diagnostic categories and hospital admission routes, and to determine whether these differences were altered following the declaration of the COVID-19 pandemic.

Search strategy

This systematic review aimed to identify studies that examined the impact of the COVID-19 pandemic on hospital admissions, with a specific focus on disaggregation by sex. The search strategy was organized around three conceptual areas: the COVID-19 pandemic, sex or gender analysis, and hospital admissions.

Electronic searches were performed in Web of Science (https://www.webofscience.com/wos/), PubMed (https://pubmed.ncbi.nlm.nih.gov/), and Cochrane Library (https://www.cochranelibrary.com/advanced-search). Reference lists of included studies were also manually screened for additional relevant sources.

The search string incorporated the following designated terms:

• i. Terms related to the COVID-19 pandemic:

  • Synonyms and variants were included (COVID-19 OR SARS-CoV-2 OR pandemic OR pandemics). To exclude studies focused on patients with persistent COVID-19 or on comparisons between infected and uninfected individuals, Boolean operators were used: NOT (long COVID OR Post-Acute COVID-19 Syndrome OR with COVID OR with and without COVID).
• ii. Sex and gender-related terms:

  • Broad descriptors were applied to capture relevant analyses (sex OR gender).
• iii. Hospital admission-related terms:

  • Various terminologies were used to encompass relevant outcomes (admission OR patient admission OR hospital OR discharge* OR patient discharge).The search was conducted in March 2024. The complete search string is provided in the supplementary material.

Eligibility criteria

Studies published from 2020 to 2024 were considered eligible if they: (1) analyzed indicators of hospital admissions; (2) reported findings by sex/gender; (3) defined both pre-pandemic and pandemic periods; and (4) were original articles published in English or Spanish.

Exclusion criteria included studies that: (1) focused on patients with COVID-19, long COVID, or comparisons between patients with and without COVID-19; (2) were conducted in older adult care centers; (3) were not empirical studies; (4) examined only a single time period (pre-pandemic or pandemic), did not define the period, or analyzed less than one year; (5) lacked explanation of statistical methods or study design; or (6) did not conduct sex-based comparisons between pre-pandemic and pandemic phases.

Study selection

Due to the high volume of search results, a structured screening process was conducted based on PRISMA flowchart guidelines [19]. Three researchers (N.V., A.L., H.A.) independently reviewed titles, abstracts, and full texts using a predefined procedure.

Duplicates were removed, followed by an initial screening of titles and abstracts. Studies were excluded at this stage for the following reasons: (i) the study population included only patients with confirmed COVID-19 or long COVID; (ii) comparative analysis focused on patients with and without COVID-19; (iii) no relevance to hospital care indicators; (iv) conducted in geriatric or long-term care facilities; or (v) not empirical.

A second phase of full-text screening was performed by two independent reviewers (I.Z., M.K.), applying additional exclusion criteria: (i) absence of clearly defined pre-pandemic and pandemic periods; (ii) comparative periods shorter than one year; (iii) focus on a single time period (pre-pandemic or during the pandemic); (iv) inadequate description of statistical methods; and (v) lack of sex-disaggregated comparative analysis.

Data extraction

From each included article, the following data were systematically extracted: study design, population characteristics, participant age, definitions of the pre-pandemic and pandemic periods, diagnostic classification system used, and primary findings relevant to the objectives. As the International Classification of Diseases, Tenth Revision (ICD-10) was the most common diagnostic system across the studies, the results are presented according to ICD-10 chapters to facilitate harmonization and comparability across heterogeneous settings.

Potential effect modifiers and reasons for heterogeneity

Although a meta-analysis was initially considered, it was ultimately deemed unfeasible due to significant methodological heterogeneity. Differences across studies in sample size, geographic setting, populations studied, definitions of pandemic periods, diagnostic classification systems, and the extent of sex-disaggregated results precluded the generation of valid pooled estimates. Therefore, this review adopts a systematic approach to describe and interpret findings within their respective contexts.

Study quality assessment

This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to ensure a comprehensive and evidence-based presentation of key elements [19]. The Joanna Briggs Institute critical appraisal checklist for analytical cross-sectional studies was used to evaluate the methodological quality of the included studies. Only primary studies classified as good quality were included.

Results

Descriptive summary

The initial search yielded 7,001 articles. After removing duplicates, 6,921 records remained. Based on the criteria outlined in the study selection section, 6,880 articles were excluded during title, abstract, and full-text screening. Ultimately, 41 articles met the inclusion criteria and were retained for final review (Fig. 1). The synthesis of all selected articles can be seen in Table 1.

Fig. 1.

PRISMA flowchart of study selection

Table 1.

Results of comparisons between men and women for different diseases in the pre-pandemic and during pandemic period

Author (year)	1) Sample size (total sample size, pre-pandemic and during pandemic) 2) Population 3) Age in years	Definition pre-pandemic and during pandemic	Type of diagnostic classification (if not specified: no standardised classification). Diagnosis.	Results • Percentage of women (w) and men (m) in the pre-pandemic and during pandemic period. P value. • Trend within each sex. • If another analysis has been performed by sex
Injury, poisoning and certain other consequences of external causes
Barandaran-Binazir et al. (2023) [20]	1) N = 5,014 (pre-pandemic n = 4,241; during-pandemic n = 773). 2) Iran. Teheran. Sina hospital. 3) All ages.	Pre-pandemic: 25/07/2016-18/02/2020. During-pandemic: 19/02/2020-10/03/2021.	ICD-10 Trauma mechanisms, blunt trauma and others trauma (drowning, poisoning, animal bite…)	• Significant differences from pre-pandemic (w: 15.1%; m: 84.9%) to during pandemic (w: 10.3%; m: 89.7%); p < 0.001.
Demir, Kircicegi, & Okut (2023) [21]	1) N = 203 (pre-pandemic n = 88; during-pandemic n = 115). 2) Turkey. One clinic. Traumatised patients admitted to the emergency department and undergoing immediate surgical procedures within the general surgery clinic. 3) ≥ 18 years.	Pre-pandemic: 10/2018-03/2020. During-pandemic: 03/2020-07/2021.	No standardised classification. Traumatised (requiring surgical intervention).	• No significant differences from pre-pandemic (w: 8.9%; m: 91.1%) to during pandemic (w: 4.9%; m: 95.1%); p = 0.109.
Domínguez et al. (2022) [22]	1) N = 1,232 (pre-pandemic n = 689; during-pandemic n = 543). 2) Australia. Melbourne. Hospital St Vincent. 3) ≥ 18 years	Pre-pandemic: 30/07/2018-15/03/2020. During-pandemic: 16/03/2020-30/10/2021.	ICD-10 Traumatic brain injury related to assault (S00–S09.2 and S09.7–S09.9).	• No significant differences from pre-pandemic (w: 17.6%; m: 82.4%) to during pandemic (w: 17.5%; m: 82.5%); p > 0.05.
Gallaher et al. 2022 [23]	1) N = 137,867 (pre-pandemic n = 124,438; during-pandemic n = 13,526). 2) Malawi. Lilongwe. Kamuzu Central Hospital. 3) Pre-pandemic mean age = 24.9 (SD: 24.8, 25.0) years; during-pandemic mean age = 28 years (SD: 27.8, 28.3).	Pre-pandemic: 01/2011-10/03/2020. During-pandemic: 11/03/2020-06/2021	No standardised classification. Traumatic injuries (excluding burn victims)	• Significant differences from pre-pandemic (w: 26%; m: 74%) to during pandemic (w: 21%; m: 79%); p < 0.001.
Frontera et al. (2022) [24]	1) N = 17,346 (pre-pandemic n = 10,567; during-pandemic n = 6,779). 2) Puerto Rico. Carolina. Hospital Dr. Federico Trilla. 3) All years.	Pre-pandemic: 2019. During-pandemic: 2020.	ICD-10 Musculoskeletal (MSK) system and connective tissue and injury or other specific consequences of external causes.	• Significant differences from pre-pandemic (w: 53.6%; m: 46.4%) to during pandemic (w: 51.4%; m: 48.6%); p = 0.007. • The median number of MSK daily encounters was significant differences from pre-pandemic (w: 16 (IQR: 13–18); m: 13 (IQR: 11–16)) to during pandemic (w: 9 (IQR: 7–12); m: 9 (IQR: 6–12)); p < 0.001.
Kontos et al. (2021) [25]	1) N = 3,021 (pre-pandemic n = 1,882; during-pandemic n = 1,139). 2) Pennsylvania. Pittsburgh. 6 clinics. 3) All ages.	Pre-pandemic: 03/2019-02/2020. During-pandemic: 03/2020-02/2021.	No standardised classification. Concussion	• No significant differences from pre-pandemic (w: 55%; m: 45%) to during pandemic (w: 55.1%; m: 44.9%); OR 0.99 IC del 95% (0.86–1.16).
Lotan et al. (2022) [26]	1) N = 172 (pre-pandemic n = 134; during-pandemic n = 38). 2) Israel. Holon. Wolfson Medical Centre. 3) > 65 years.	Pre-pandemic: 2018–2019. During-pandemic: 2020.	No standardised classification Acute vertebral compression fracture (VCF).	• No significant differences from pre-pandemic (w: 71.6%; m: 28.4%) to during pandemic (w: 65.8%; m: 34.2%); p = 0.46.
Philip et al. (2022) [27]	1)N = 2,600 (pre-pandemic n = 1,691; during-pandemic n = 909). 2) India. Kerala. Kottayam. Government Medical College and Hospital. 3) From 10 months to 98 years	Pre-pandemic: 01/02/2019-31/01/2020. During-pandemic: 01/02/2020-31/01/2021.	No standardised classification. Maxillofacial fracture.	• No significant differences from pre-pandemic to during pandemic for sex; p = 0.72. Classification by causes: • Facial fractures caused by traffic accidents: pre-pandemic (w: 48.6%; m: 68.3%); during-pandemic (w: 53.1%; m: 70.1%). • Facial fractures caused by falls at home: pre-pandemic (w: 20.6%; m: 7%); during-pandemic: (w: 30.9%; m: 14.9%). • Facial fractures in accidents with multiple victims: pre-pandemic (w: 9%; m: 2.5%). • Facial fractures because of occupational hazard, fall from height and sports injury in pre-pandemic period not be significant between males and females. Classification by location: • Frontal bone fractures: pre-pandemic (w: 5.5%; m: 12.8%); during-pandemic (w: 8.6%; m: 16.5%). • Fractures of primary teeth: pre-pandemic (w: 1.3%; m: 0.2%); during-pandemic (w: 2.5%; m: 0.1%). • Dentoalveolar fractures: pre-pandemic (w: 50.8%; m: 38.6%). • Nasal fractures: pre-pandemic: (w: 3.1%; m: 14.3%).
Basu and Alex (2023) [28]	1) N = 779 (pre-pandemic n = 462; during-pandemic n = 317). 2) India. Calcutta. Department of Plastic and Reconstructive Surgery, Burns Unit. 3) All ages.	Pre-pandemic: 09/2018-03/2020. During-pandemic: 03/2020-08/2021.	No standardised classification. Acute burns	• Significant differences from pre-pandemic (w: 57.4%; m: 42.6%) to during pandemic (w: 74.1%; m: 25.9%); p = 0.00001.
Sung et al. (2023) [29]	1) N = 5,073 (pre-pandemic n = 971; during-pandemic n = 663). 2) Korea. Wonju Severance Christian Hospital. 3) Pre-pandemic media 58.3 years (SD: ± 20.4); during-pandemic media 59.6 years (SD: ± 18.7).	Pre-pandemic: 01/2018-12/2019. During-pandemic: 01/2020-12/2021.	No standardised classification. Traumatic brain injury	• No significant differences from pre-pandemic (w: 26.6%; m: 73.4%) to during pandemic (w: 26.1%; m: 73.9%); p = 0.830.
Wong et al. (2024) [30]	1) N = 657 (pre-pandemic n = 307; during-pandemic n = 350). 2) Singapore. Tertiary institution. 3) ≥ 60 years.	Pre-pandemic: 2019. During-pandemic: 2020.	No standardised classification. Surgery for fractures of the femoral neck and peritrochanteric fractures.	• No significant differences from pre-pandemic (w: 69.4%; m: 30.6%) to during pandemic (w: 66.3%; m: 33.7%); p = 0.397.
Mental and behavioural disorders
Beaudry et al. (2022) [31]	1) N = 9,748 (pre-pandemic n = 5,927; during-pandemic n = 3,821). 2) Canada. Montreal. Centre hospitalier universitaire. Sainte-Justine. 3) 5–17 years.	Pre-pandemic: 01/04/2016-26/02/2020. During-pandemic: 01/03/2020-30/11/2021.	No standardised classification. Pathology related to mental health.	• Significant differences from prepandemic (w: 59.5%; m: 40.6%) to during pandemic (2020 w: 63.8%; m: 36.2%; 2021 w: 70.2%; 29.8%); p < 0.001.
Fluck et al. (2023) [32]	1) N = 21,192 (pre-pandemic n = 10,173; during-pandemic n = 11,019). 2) UK. Surrey. One hospital. 3) ≥ 18 years.	Pre-pandemic: 01/04/2019–29/02/2020. During-pandemic: 01/03/2020–31/03/2021.	ICD-10 Mental illness (F10-F99) and self-poisoning (T36-T65).	• No significant differences from pre-pandemic (w: 52.3%; m: 47.7%) to during pandemic (w: 52.5%; m: 47.5%); p= 0.358. • Higher admission rates for mental health disorders and self-poisoning during the pandemic in both sexes. • No significant differences in CCI (score ≥ 2) between periods of study in either sex.  o W pre-pandemic: 1.5%; during-pandemic: 1.8%; p= 0.389.  o M pre-pandemic: 1.9%; during-pandemic: 2.4%; p= 0.186. • Significant differences in men with mental health disorders and a CCI score ≥ 2:  o W pre-pandemic: 0.7%; during-pandemic: 0.9%; p= 0.109.  o M pre-pandemic: 0.8%; during-pandemic: 1.2%; p= 0.032. • Significant differences in women with self-inflicted poisoning and a CCI score ≥ 2:  o W pre-pandemic: 1.8%; during-pandemic: 2.4%; p= 0.015.  o M pre-pandemic: 1%; during-pandemic: 1.3%; p= 0.121. • Significant differences in admission rates due to self-poisoning; higher in women (2.1%) than in men (1.2%): (p < 0.001).
Marchili et al. (2024) [37]	1) N = 260 (pre-pandemic n = 36; during-pandemic n = 224). 2) Italy. Rome. Bambino Gesù Children’s Hospital. 3) 6 to 18 years.	Pre-pandemic: 03/2019-02/2020. During-pandemic: 03/2020-10/2022.	DSM V Anorexia nervosa	• No significant differences from pre-pandemic (w: 86.1%; m: 13.9%) to during pandemic (w: 94.6%; m: 5.4%); p = 0.0681.
Morkavuk et al. (2021) [33]	1) N = 202 (pre-pandemic n = 128; during-pandemic n = 74). 2) Turkey. 3) 55–96 years.	Pre-pandemic: 03/2019-02/2020. During-pandemic: 03/2020-02/2021.	No standardised classification. Dementia.	• No significant differences from pre-pandemic (w: 58.6%; m: 41.4%) to during pandemic (w: 70.3%; m: 29.7%); p = 0.098.
Peraire et al. (2022) [34]	1) N = 1,579 (pre-pandemic n = 785; during-pandemic n = 794). 2) Spain. Castellon. Short-Stay Hospitalisation Unit of the Provincial Hospital Consortium. 3) 18 to 92 years	Pre-pandemic: 04/2018- 02/2020. During-pandemic: 02/2020-12/2021.	DSM-IV and DSM-V Psychiatric pathology.	• No significant differences from pre-pandemic (w: 45.5%; m: 54.5%) to during pandemic (w: 43.6%; m: 56.4%); p = 0.447.
Diseases of the digestive system (including gastrointestinal bleeding, alcoholic liver disease, and appendicitis)
Campbell et al. (2023) [35]	1) N = 1,378 (pre-pandemic n = 678; during-pandemic n = 700). 2) USA. Hospital admissions. 3) Mean age 53.9 +/-13.1.	Pre-pandemic: 04/2019-03/2020. During-pandemic: 04/2020-03/2021.	ICD-10 Alcoholic liver disease (K70 and its sub-categories: the spectrum of alcoholic fatty liver, hepatitis, cirrhosis, fibrosis).	• Significant differences from pre-pandemic (w: 30.4%; m: 69.6%) to during pandemic (w: 38.1%; m: 61.9%); p = 0.002. • Significant increase in the number of hospital admissions due to ALD in women (33%) and men under 50 (24%) from the pre-pandemic to the pandemic period. • Significant increase in the number of younger women requiring admission for ALD was identified. • Decrease in ALD hospitalizations for older male
Kwei-Nsoro et al. (2023) [36]	1) N = 927,014 (pre-pandemic n = 486,780; during-pandemic n = 440,234). 2) USA. Administrative data from the National Inpatient Sample (NIS) for the years 2016–2020. 3) ≥ 18 years.	Pre-pandemic: 2019. During-pandemic: 2020.	ICD-9 Gastrointestinal bleeding.	• Admitted with a primary diagnosis of gastrointestinal bleeding: significant differences from pre-pandemic (w: 48.62%; m: 51.38%) to during pandemic (w: 48.04%; m: 51.96%); p = 0.027. • Mortality of admitted adults: no significant differences from woman (pre-pandemic: 1.9; during-pandemic: 2.08) p = 0.059; significant differences from male (pre-pandemic: 2.08; during-pandemic: 2.41) p = 0.007. • Average length of hospital stay in days from woman (pre-pandemic: 4.36 ± 0.03; during-pandemic: 4.41 ± 0.03); from male (pre-pandemic: 4.29 ± 0.03; during-pandemic: 4.36 ± 0.03). • Total costs of hospital admissions from woman (pre-pandemic: 50,563 ± 668; during-pandemic: 51,612 ± 675); from male (pre-pandemic: 53,994 ± 740; during-pandemic: 55,194 ± 708).
Mohajerzadeh et al. (2023) [37]	1) N = 369 (pre-pandemic n = 173; during-pandemic n = 196). 2) Iran. Tehran. Mofid Children’s Hospital. 3) < 14 years.	Pre-pandemic: 02/2019-02/2020. During-pandemic: 02/2020-02/2021.	ICD-10 Acute appendicitis.	• No significant differences from pre-pandemic (w: 35.5%; m: 64.5%) to during pandemic (w: 35.8%; m: 64.2%); p = 0.955.
Sener Okur et al. (2022) [38]	1) N = 285 (pre-pandemic n = 142; during-pandemic n = 143). 2) Turkey. Denizili. State hospital. Paediatric Surgery Department. 3) Paediatric. Not specified	Pre-pandemic: 18/03/2019-18/03/2020. During-pandemic: 18/03/2020-18/03/2021.	No standardised classification. Acute appendicitis.	• Significant differences from pre-pandemic (w: 30.3%; m: 69.7%) to during pandemic (w: 46.9%; m: 53.1%); p = 0.01.
Zheng et al. (2021) [39]	1) N = 1,135 (pre-pandemic n = 652; during-pandemic n = 483). 2) China. Beijing. Hospital Jishuitan. 3) Mean age pre-pandemic: 37.17 (SD: ± 0.58); during-pandemic 38.59 years (SD: ± 0.71).	Pre-pandemic: 2019. During-pandemic: 2020.	No standardised classification. Acute appendicitis.	• No significant differences from pre-pandemic (w: 48.2%; m: 51.8%) to during pandemic (w: 45.8%; m: 54.2%); p = 0.422.
Diseases of the circulatory system
Biel et al. (2024) [40]	1) N = 262 (pre-pandemic n = 138; during-pandemic n = 124). 2) Poland, Lower Silesia. 3) 20–90 years.	Pre-pandemic: 01/01/2018-31/12/2019. During-pandemic: 04/03/2020-03/03/2022.	No standardised classification. ICD/CRT-D shock	• No significant differences from pre-pandemic (w: 17.4%; m: 82.6%) to during pandemic (w: 17.7%; m: 82.3%); p = 0.94.
Duhan et al. (2023) [41]	1) N = 1,022 (pre-pandemic n = 511; during-pandemic n = 511) 2) USA. Maryland state. Baltimore. Sinai Hospital. 3) Between 52–81 years	Pre-pandemic: 01/01/2018–20/01/2020. During-pandemic: 21/01/2020–31/12/2021.	No standardised classification. Post-percutaneous coronary intervention and adverse events.	• Differences from pre-pandemic (w: 38.2%; m: 61.8%) to during pandemic (w: 37.4%; m: 62.3%). • Non AA: pre-pandemic (w: 33%; m: 67%) to during pandemic (w: 30%; m: 70%). p > 0.05 • AA: pre-pandemic (w: 45%; m: 55%) to during pandemic (w: 47%; m: 53%). p > 0.05 • AA women experienced higher rates of ischaemic events during the pandemic compared to other genders and races. • In a gender-based analysis, no statistical difference in any ischemic events was found within the same race-gender groups before vs. during COVID-19 or within the same period-gender category among AAs vs. non-AAs. • Before the pandemic, AA women had numerically the highest ischemic event occurrences with non-AA men before pandemic as a baseline (OR: 1.7, 95% CI: 0.87–3.34, non-significant). • During-pandemic, the highest rate of ischemic events was seen in AA women (OR: 3.58, 95% CI: 1.89–6.81, p = 0.0001) followed by AA men (OR: 2.51, 95% CI: 1.31–4.82, p = 0.005) and non-AA women (OR:3.58, 95% CI: 1.89–6.81, p = 0.0001).
Lopez-de-Andres et al. (2023) [42]	1) N = 45,742 (pre-pandemic n = 24,675; during-pandemic n = 21,067). 2) Spain. National Health System Hospital Admissions Database. 3) ≥ 18 years.	Pre-pandemic: 01/01/2019-31/12/2019. During-pandemic: 01/01/2020-31/12/2020.	ICD-10 Patients with DM who have undergone any of the following cardiac procedures: CABG; PCI; OSVR; TCVI.	• CABG: No significant differences from pre-pandemic (w: 19.2%; m: 80.8%) to during pandemic (w: 18.7%; m: 81.3%); p = 0.651. • ICP: No significant differences from pre-pandemic (w: 26.5%; m: 73.5%) to during pandemic (w: 25.9%; m: 74.1%); p = 0.221. • OSVR: No significant differences from pre-pandemic (w: 37.7%; m: 62.3%) to during pandemic (w: 36.6%; m: 63.4%); p = 0.461. • TCVI: No significant differences from pre-pandemic (w: 47.4%; m: 52.6%) to during pandemic (w: 48.1%; m: 51.9%); p = 0.681.
Vasaghi Gharamaleki et al. (2022) [43]	1) N = 127 (pre-pandemic n = 50; during-pandemic n = 77). 2) Iran. Shiraz. Namazi Hospital. 3) > 20 years.	Pre-pandemic: 21/03/2018-20/03/2019. During-pandemic: 20/03/2020-20/03/2021.	ICD-10 G08: Intracranial and intraspinal phlebitis and thrombophlebitis. O87.3: Cerebral venous thrombosis in the puerperium. I63.6 Cerebral infarction due to cerebral venous thrombosis, nonpyogenic. I67.6 Non-pyogenic thrombosis of the intracranial venous system. I61.9 Non-traumatic intracerebral hemorrhage.	• No significant differences from pre-pandemic (w: 76%; m: 24%) to during pandemic (w: 67.5%; m: 32.5%); p = 0.305. • The prognosis at three months of follow-up was not associated with gender in the pre-pandemic period or during the pandemic.
Diseases of the genitourinary system
Erkan et al. (2023) [44]	1) N = 2,018 (pre-pandemic n = 1242; during-pandemic n = 776). 2) Turkey. Bursa Yüksek İhtisas Training and Research Hospital. 3) All years.	Pre-pandemic: 10/03/2019-09/03/2020. During-pandemic: 10/03/2020-10/03/2021.	No standardised classification. Urological pathology.	• No significant differences from pre-pandemic (w: 22.6%; m: 77.4%) to during pandemic (w: 22.8%; m: 77.2%); p > 0.05.
Gul et al. (2022) [45]	1) N = 250 (pre-pandemic n = 182; during-pandemic n = 68). 2) Turkey. Urology Polyclinic. 3) < 18 years.	Pre-pandemic: 11/03/2019-11/03/2020. During-pandemic: 11/03/2020-11/03/2021.	No standardised classification. UTI symptoms and had bacterial growth in their urine cultures	• Significant differences from pre-pandemic (w: 84.1%; m: 15.9%) to during pandemic (w: 67.6%; m: 32.4%); p = 0.004. • Distribution of the bacteria:  o E. coli: No significant differences from pre-pandemic (w: 91.6%; m: 8.4%) to during pandemic (w: 80%; m: 20%); p = 0.166.  o K. pneumoniae: No significant differences from pre-pandemic (w: 55.6%; m: 44.4%) to during pandemic (w: 85.7%; m: 14.3%); p = 0.158.  o P. mirabilis: No significant differences from pre-pandemic (w: 50%; m: 50%) to during pandemic (w: 50%; m: 50%); p = 0.99.  o Others: Significant differences from pre-pandemic (w: 90%; m: 10%) to during pandemic (w: 29.4%; m: 70.6%); p = 0.002.
Endocrine, nutritional and metabolic diseases
Popa et al. (2023) [46]	1) N = 1,505 (pre-pandemic n = 1,055; during-pandemic n = 450). 2) Romania. Timisoara County Hospital ‘Pius Brinzeu’. 3) Not specifies.	Pre-pandemic: 01/2018-02/2020. During-pandemic: 03/2020-12/2021.	No standardised classification. Nodular thyroid disease treated surgically.	• No significant differences from pre-pandemic (w: 90.05%; m: 9.95%) to during pandemic (w: 88.4%; m: 11.6%); p = 0.36. • There were no statistically significant differences in malignant tumours by sex and age group in the pre-pandemic period and during the pandemic.
Sekowski et al. (2022) [47]	1) N = 114,065 (pre-pandemic n = 68,906; during-pandemic n = 45,159). 2) Poland. Public and private hospitals (except psychiatric units) in all administrative regions. 3) All years.	Pre-pandemic: 2019. During-pandemic: 2020.	ICD-10 Different types of DM. E10: insulin-dependent diabetes mellitus; E11: non-insulin-dependent diabetes mellitus; E13: other specified diabetes mellitus; E14: unspecified diabetes mellitus.	Hospitalization rate per 100,000 inhabitants  • Overall DM: Significant differences from hospitalization rate per 100,000 inhabitants pre-pandemic (w: 163.7; m: 196.4) to during pandemic (w: 102.7; m: 134.3); (% difference pre-pandemic to during pandemic (w: -37.2%; m: -31.6); p < 0.001.  • E10: Significant differences from hospitalization rate per 100000 inhabitants pre-pandemic (w: 64.4; m: 85.4) to during pandemic (w: 44.1; m: 62.5); (% difference pre-pandemic to during pandemic (w: -37.2%; m: -26.7%); p < 0.001.  • E11: Significant differences from hospitalization rate per 100000 inhabitants pre-pandemic (w: 95; m: 104) to during pandemic (w: 56.1; m: 67.5); (% difference pre-pandemic to during pandemic (w: -40.9%; m: -35.2%); p < 0.001.  • E13: No significant differences from hospitalization rate per 100000 inhabitants pre-pandemic (w: 0.3; m: 0.6) to during pandemic (w: 0.2; m: 0.4); (% difference pre-pandemic to during pandemic (w: -37.1%; m: -38.4%); p = 0.9.  • E14: No significant differences from hospitalization rate per 100000 inhabitants pre-pandemic (w: 0.3; m: 0.6) to during pandemic (w: 0.2; m: 0.4); (% difference pre-pandemic to during pandemic (w: -37.1%; m: -38.4%); p = 0.9.
				Hospital mortality rate (%): Type I diabetes  • All years: Significant differences from pre-pandemic (w: 2.3; m: 2) to during pandemic (w: 3.8; m: 3.2); (% difference pre-pandemic to during pandemic (w: 65.2%; m: 60%); p < 0.001.  • 40–49 years: Significant differences from pre-pandemic (w: 0.4; m: 1.3) to during pandemic (w: 1; m: 2.1); (% difference pre-pandemic to during pandemic (w: 150%; m: 61.5%); p = 0.03.  • 50–59 years: No significant differences from pre-pandemic (w: 1.3; m: 2.2) to during pandemic (w: 2; m: 3.1); (% difference pre-pandemic to during pandemic (w: 53.8%; m: 40.9%); p = 0.052.  • 60–69 years: Significant differences from pre-pandemic (w: 3.5; m: 2.7) to during pandemic (w: 4.1; m: 4.7); (% difference pre-pandemic to during pandemic (w: 17.1%; m: 74.1%); p < 0.001.  • 70–79 years: Significant differences from pre-pandemic (w: 4.9; m: 5.2) to during pandemic (w: 9.6; m: 8); (% difference pre-pandemic to during pandemic (w: 95.9%; m: 53.8%); p < 0.001.  • > 80 years: Significant differences from pre-pandemic (w: 9.7; m: 9.8) to during pandemic (w: 14.7; m: 13.2); (% difference pre-pandemic to during pandemic (w: 51.5%; m: 34.7%); p < 0.001.
				Type II diabetes  • All years: Significant differences from pre-pandemic (w: 3.8; m: 2.9) to during pandemic (w: 5.4; m: 4.5); (% difference pre-pandemic to during pandemic (w: 42.1%; m: 55.2%); p < 0.001.  • 40–49 years: No significant differences from pre-pandemic (w: 0.9; m: 1) to during pandemic (w: 0.5; m: 1.2); (% difference pre-pandemic to during pandemic (w: -44.4%; m: 20%); p = 0.09.  • 50–59 years: Significant differences from pre-pandemic (w: 0.9; m: 1.2) to during pandemic (w: 2.3; m: 1.8); (% difference pre-pandemic to during pandemic (w: 155.6%; m: 50%); p = 0.001.  • 60–69 years: Significant differences from pre-pandemic (w: 1.9; m: 2.2) to during pandemic (w: 3.3; m: 3.5); (% difference pre-pandemic to during pandemic (w: 73.7%; m: 59.1%); p < 0.001.  • 70–79 years: Significant differences from pre-pandemic (w: 3.4; m: 3.7) to during pandemic (w: 4.7; m: 5.9); (% difference pre-pandemic to during pandemic (w: 38.2%; m: 59.5%); p < 0.001.  • > 80 years: Significant differences from pre-pandemic (w: 7.5; m: 7.8) to during pandemic (w: 9.4; m: 11.8); (% difference pre-pandemic to during pandemic (w: 25.3%; m: 51.3%); p < 0.001. Throughout the period analysed (2014–2020), the overall rate of hospitalisation for type 1 diabetes was higher among men than among women.
Neoplasms
Kleemann et al. (2022) [48]	1) N = 248,479 (pre-pandemic n = 133,771; during-pandemic n = 114,708). 2) Germany. All German hospitals. 3) Not specified (< 65; 65–74; 75+).	Pre-pandemic: 18/03/2019-17/03/2020. During-pandemic: 18/03/2020-17/03/2021.	ICD-10 and OPS codes. Skin cancer. Melanomas and non-melanomas.	• Malignant melanoma: significant differences from pre-pandemic (w: 44.4%; m: 55.6%) to during pandemic (w: 45.5%; m: 54.5%); p = 0.0410. • Non-melanoma skin cancer: significant differences from pre-pandemic (w: 41.6%; m: 58.4%) to during pandemic (w: 41.1%; m: 58.9%); p = 0.0016. • Melanoma in situ: no significant differences from pre-pandemic (w: 49.9%; m: 50.1%) to during pandemic (w: 51.4%; m: 48.6%); p = 0.8169. • Carcinoma in situ of the skin: no significant differences from pre-pandemic (w: 46.2%; m: 53.8%) to during pandemic (w: 44.7%; m: 55.3%); p = 0.2937.
Pages et al. (2021) [49]	1) N = 36,014 (pre-pandemic n = 24,380; during-pandemic n = 11,634). 2) France. Public and private hospitals. 3) ≥ 18 years.	Pre-pandemic:2018–2019. During-pandemic: 2020.	ICD-10 Lung cancer. (ICD-10 code C34 and a lung resection procedure during hospitalisation).	• Significant differences 2018 (w: 36.6%; m: 63.4%); 2019 (w: 38%; m: 62%); 2020 (w: 39.4%; m: 60.6); p < 0.0001. • Significant differences in non-COVID patients hospitalised for lung cancer resection between 2019 (m: 62%) and 2020 (m: 60.6%); p = 0.0204.
External causes of morbidity and mortality (sexual assault, falls, self-harm)
Arslan et al. (2023) [50]	1) N = 310,848 (pre-pandemic n = 190,540; during-pandemic n = 120,308). 2) Turkey. Emergency department of the Ege University Faculty of Medicine Hospital. 3) ≥ 65 years.	Pre-pandemic: 11/03/2019-10/03/2020. During-pandemic: 11/03/2020-10/03/2021.	ICD-10 Falls (W00-19).	• No significant differences from pre-pandemic (w: 61%; m: 39%) to during pandemic (w: 62.6%; m: 37.4%); p = 0.34.
Otani et al. (2023) [51]	1) N = 201 (pre-pandemic n = 100; during-pandemic n = 101). 2) Japan. Kobe. General Hospital of the Medical Centre. 3) 13–90 years.	Pre-pandemic: 01/04/2018-31/03/2020. During-pandemic: 01/04/2020-31/03/2022.	No standardised classification. Suicide attempts.	• No significant differences from pre-pandemic (w: 61%; m: 39%) to during pandemic (w: 57.4%; m: 42.6%); p = 0.667.
Saxena et al. (2023) [52]	1) N = 42,202 (pre-pandemic n = 27,839; during-pandemic n = 14,362). 2) Canada. 3) ≥ 5 years.	Pre-pandemic:2018–2019. During-pandemic: 2020.	No standardised classification. Intentional injuries (self-harm, violence, assault and child abuse).	• Intentional injuries differences from pre-pandemic (w: 50.4%; m: 49.6%) to during pandemic (w: 55%; m: 45%). • Self-harm differences from pre-pandemic (w: 68.8%; m: 31.2%) to during pandemic (w: 70.5%; m: 29.5%).  o 5–10 year significant differences patients with from pre-pandemic (w: 42.9%; m: 57.1%) to during pandemic (w: 42.5%; m: 57.5%); p < 0.0001.  o 11–18 years: differences from pre-pandemic (w: 76.1%; m: 23.9%) to during pandemic (w: 77.3%; m: 22.7%).  o 19–29 years: differences from pre-pandemic (w: 64.9%; m: 38.1%) to during pandemic (w: 55.2%; m: 44.8%).    o 30–64 years: differences from pre-pandemic (w: 45.6%; m: 54.4%) to during pandemic (w: 45.3%; m: 54.7%). o ≥ 65 years: differences from pre-pandemic (w: 37.9%; m: 62.1%) to during pandemic (w: 42.3%; m: 57.7%).
Seng et al. (2023) [53]	1) N = 1,557 (pre-pandemic n = 1,075; during-pandemic n = 482). 2) Pennsylvania. Trauma centres. 42 trauma centres verified by the Pennsylvania Trauma System Foundation. 3) ≥ 18 years.	Pre-pandemic: 01/01/2018-31/03/2020. During-pandemic: 01/04/2020-31/03/2021.	ICD-10 Self-inflicted injuries (X71, X72, X73, X74, X75, X76, X77, X78, X79, X80, X81, X82 and X83).	• No significant differences from pre-pandemic (w: 26.8%; m: 73.2%) to during pandemic (w: 28.2%; m: 71.8%); p = 0.55. • No significant differences for mortality from self-inflicted injury during the pandemic: mortality odds ratio (95% CI) for females OR 0.80 (0.39; 1.64): p  = 0.54.
Wong et al. (2022) [54]	1) N = 442,527 (pre-pandemic n = 304,390; during-pandemic n = 138,137) 2) China. Hong Kong. 18th ED 3) < 18 years	Pre-pandemic: 01/01/2016-31/12/2019. During-pandemic: 01/01/2020-31/12/2021.	ICD-9 Sexual abuse (Code: 995.53).	• Child sexual abuse was identified in 455 encounters of 427 patients: 414 cases involving 387 girls (90.6%) vs. 41 cases involving 40 boys (9.4%). • Incidence rate per million girls significantly increased (monthly mean, 9.7 pre-pandemic cases vs. 16.28 pandemic cases; RIR, 1.68; 95% CI, 1.32–2.13; p < 0.001), mostly among girls aged 12 to 17 years. • There was no significant change in CSA cases in boys per million (monthly mean, 1.16 pre-pandemic cases vs. 0.99 pandemic cases; RIR, 0.86; 95% CI, 0.43–1.70; p = 0.66;). • Although the number of CSA cases increased throughout the 24-month pandemic period, an increase often occurred immediately after resumption of face-to-face classes in secondary schools.
Various pathologies from different chapters of ICD-10
Cozzi et al. (2022) [55]	1) N = 112,168 (pre-pandemic n = 96,645; during-pandemic n = 15,523). 2) Italy. Trieste. Children’s Hospital, Institute of Maternal and Child Health, IRCCS Burlo Garofolo in Trieste. 3) 0–17 years.	Pre-pandemic: 03/2016-02/2020. During-pandemic: 03/2020-02/2021.	No standardised classification. Respiratory disease, neurological disease, injury, psychiatric disease, cardiovascular disease, non-respiratory infections, poisoning and others.	• No significant differences from pre-pandemic (w: 39%; m: 61%) to during pandemic (w: 45.2%; m: 54.8%); p = 0.453.
Kim et al. (2022) [56]	1) N = 2,808,756 (pre-pandemic n = 1,835,045; during-pandemic n = 973,711). 2) Korea. 402 emergency services nationwide. 3) 0–18 years.	Pre-pandemic: 01/2019-12/2019. During-pandemic: 01/2020-12/2020.	ICD-10 Infectious disease, mental health disorder, injury, self-harm or suicide attempts, violence or aggression	• No significant differences from pre-pandemic (w: 43.5%; m: 56.5%) to during pandemic (w: 43.4%; m: 56.6%); p = 0.29. • Infectious disease: Significant differences from pre-pandemic (w: 45.9%; m: 54.1%) to during pandemic (w: 46.8%; m: 53.2%); p < 0.01. • Mental health disorders: Significant differences from pre-pandemic (w: 65.2%; m: 34.8%) to during pandemic (w: 68.3%; m: 31.7%); p < 0.01. • Injury: Significant differences from pre-pandemic (w: 37.7%; m: 62.3%) to during pandemic (w: 38.5%; m: 61.5%); p < 0.01.
Diseases of the skin and subcutaneous tissue
Temel et al. (2023) [57]	1) N = 639 (pre-pandemic n = 467; during-pandemic n = 172). 2) 1) Turkey. Ankara. The Dermatology Clinic of Ankara Training and Research Hospital. 3) Adults. Pre-pandemic media 44.4 years (SD: ± 18.6); During-pandemic media 46.1 years (SD: ± 18.2).	Pre-pandemic: 11/03/2019-10/03/2020. During-pandemic: 11/03/2020-11/03/2021.	ICD-10 Dermatological pathology.	• No significant differences from pre-pandemic (w: 48.8%; m: 51.2%) to during pandemic (w: 51.2%; m: 48.8%); p = 0.38.
Diseases of the eye and adnexa
Yilmaz et al. (2022) [58]	1) N = 161,941 (pre-pandemic n = 103,178; during-pandemic n = 58,763). 2) Turkey. A tertiary care ophthalmology hospital. 3) All years.	Pre-pandemic: 15/06/2019-15/03/2020. During-pandemic: 15/03/2020-15/03/2021.	No standardised classification. Ophthalmological emergency pathology.	• Significant differences from pre-pandemic (w: 36.7%; m: 63.3%) to during pandemic (w: 32.4%; m: 67.6%); p = 0.001. • Significant differences from pre-pandemic to during pandemic by age group:  o 0–16 years: from pre-pandemic (w: 44.4%; m: 55.6%) to during pandemic (w: 40.6%; m: 59.4%); p = 0.001.  o 17–40 years: from pre-pandemic (w: 33.7%; m: 66.3%) to during pandemic (w: 29.4%; m: 70.6%); p = 0.001.  o 41–64 years: from pre-pandemic (w: 33.2%; m: 66.8%) to during pandemic (w: 30.7%; m: 69.3%); p = 0.001. o No significant differences ≥ 65 years from pre-pandemic (w: 47.2%; m: 52.8%) to during pandemic (w: 47.8%; m: 52.2%); p = 0.51.
Symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified (epistaxis)
Vlaescu et al. (2023) [59]	1) N = 380 (pre-pandemic n = 231; during-pandemic n = 149). 2) Poland and Romania. Otolaryngology Clinic at Craiova County Emergency Hospital and at the Institute of Speech Therapy and Functional Otolaryngology Surgery. 3) Not specified. Classification: young adults; middle aged; old adults.	Pre-pandemic: 10/2018-02/2020. During-pandemic: 03/2020-05/2022.	No standardised classification. Epistaxis.	• No significant differences from pre-pandemic (w: 45.9%; m: 54.1%) to during pandemic (w: 45%; m: 55%); p = 0.916.

AA African American, AH Alcohol-associated hepatitis, ALD Alcoholic liver disease, CABG coronary artery bypass graft, CCI Charlson comorbidity index / Charlson comorbiity index, CRT-D Cardiac resynchronization therapy with defibrillation, DM diabetes mellitus, DSM-IV Diagnostic and Statistical Manual of Mental Disorders of the American Psychiatric Association, fourth edition, DSM V Diagnostic and Statistical Manual of Mental Disorders of the American Psychiatric Association, fifth edition, ED Emergency DepartmentI, CD Implantable cardioverter-defibrillators, ICD 9 International Classification of Diseases, Ninth Revision, ICD 10 International Classification of Diseases, Tenth Revision, IQR Interquartile Ranges, OPS Operation and procedure codes, OSVR open surgical valve replacement, PCI percutaneous coronary intervention (está duplicado — elige solo uno), RD Rate differences, RTA Road Traffic Accident, TCVI transcatheter valve implantation, UTI Urinary tract infection, VCF Vertebral compression fracture

All selected articles employed quantitative methodologies and used retrospective observational designs. All were written in English (n = 41) and conducted in Europe (n = 11), North America (n = 8), Asia (n = 20), Oceania (n = 1), and Africa (n = 1).

Sample sizes ranged from 127 to 2,808,756 participants. Participant age varied across studies, with pediatric populations (n = 8), adult populations (n = 18), and mixed-age populations (n = 15) represented. Patients were recruited from diverse settings, including hospital units, entire hospitals, and emergency departments, both at single centers and on national levels.

For diagnostic classification, some studies used standardized coding systems: ICD-9 (n = 2), ICD-10 (n = 15), and DSM-IV or DSM-V (n = 2). However, 22 studies did not specify the classification system employed.

Trauma, poisoning, and other external causes

11 studies focused on trauma, poisoning, and other consequences of external causes. Among studies involving only adults, four reported no statistically significant sex differences between the pre-pandemic and pandemic periods. In Turkey [21] and Australia [22], males experienced most injuries (surgical and craniocerebral), whereas in Israel [26] and Singapore [30] vertebral and hip fractures affected more females, with no significant changes over time.

In studies involving all age groups, a statistically significant decrease in trauma among females during the pandemic was noted in Iran for general trauma and contusions [20], in Malawi for traumatic injuries excluding burns [23] and in Puerto Rico for musculoskeletal injuries from external causes [24]. Conversely, in Korea, acute burn hospitalizations among females significantly increased during the pandemic [28].

Other studies reported no significant changes in the incidence of concussions in Pennsylvania [25], head trauma in Korea [29], and maxillofacial fractures in India [27]. However, Philip et al. (2022) noted an increased incidence of these injuries in females during the pandemic, primarily attributed to falls at home and traffic accidents.

Diseases of the digestive tract

Five studies investigated diseases of the digestive system In pediatric populations, a significant rise in hospitalizations for acute appendicitis among Turkish females was found during the pandemic [38], and a similar trend was found in Iran, although without statistical significance [37].

Two studies in adults, both conducted in the United States, found significant sex- and period-related differences. One reported increased hospitalizations for alcoholic hepatitis among females during the pandemic [35] while the other observed a significant rise in hospitalizations and mortality from gastrointestinal bleeding among males [36]. Regarding hospital stay length, females had longer stays in both periods, while males incurred higher mean hospitalization costs (pre-pandemic: 53,994; during-pandemic: 55,194) compared to females (pre-pandemic: 50,563; during-pandemic: 51,612), though no statistical analysis was provided for cost differences.

In a general population study, Zheng et al. (2021) reported a decrease in hospitalizations among females, also without statistical significance [39].

Mental and behavioral disorders

Five studies addressed mental and behavioral disorders. In pediatric populations, Beaudry et al. (2023) noted a significant increase in emergency service use among females in Canada [31]. Marchili et al. (2024) observed a rise in female hospitalizations for anorexia during the pandemic in Italy, although the difference did not reach statistical significance [60].

In adults, Fluck et al. (2023) reported significant sex differences in the United Kingdom during the pandemic: increased male admissions for mental disorders and increased female admissions for autointoxication among patients with a Charlson Comorbidity Index (CCI) score ≥ 2 [32]. In Spain and Turkey, no significant sex-based differences were found in admissions for psychiatric disorders or dementia [33, 34].

Diseases of the circulatory system

Four studies focused on diseases of the circulatory system. In the United States, Duhan et al. (2023) analyzed percutaneous coronary interventions in adults and found no significant sex differences within racial groups before and during the pandemic [41]. Also in adult populations, Biel et al. (2024) in Poland found no significant differences in implantable cardioverter-defibrillators (ICD) or cardiac resynchronization therapy with defibrillation (CRT-D) shock hospitalizations by sex or period. However, prior to the pandemic, African American females had numerically more ischemic events than other groups, with a risk 1.7 times higher than non-African American males—though this was not statistically significant. During the pandemic, their risk increased to 3.5 times higher, this time with statistical significance [40].

In Spain, Lopez-de-Andres et al. (2023) studied four cardiac procedures and observed decreased hospitalizations for coronary artery bypass grafting, open surgical valve replacement, and percutaneous coronary intervention (PCI), and increased transcatheter valve implantation in females, none of which were statistically significant [42]. In Iran, Vasaghi Gharamaleki et al. (2022) reported a reduction in cerebral venous sinus thrombosis admissions among females without statistical significance [43].

Urinary tract diseases (UTI)

Two studies focus on genitourinary disorders. In pediatric populations, Gul et al. (2022) in Turkey observed a significant decline in UTI in female hospitalizations [45]. Erkan et al. (2023), analyzing general urological pathology without age restrictions, reported no significant sex-based or temporal differences [44].

Endocrine, nutritional, and metabolic diseases

Two studies examined endocrine, nutritional, and metabolic diseases. In Romania, Popa et al. (2023) found no sex differences between periods for nodular thyroid disease across all ages [46]. In Poland, Sękowski et al., (2022) observed a significant reduction in hospitalizations among females with type 1 and type 2 diabetes mellitus (DM), alongside increased female mortality during the pandemic—except in certain age groups [47].

Neoplasms

Two studies focused on neoplasms, specifically skin and lung. Pages et al. (2021) in France reported a significant increase in hospital admissions for lung cancer with resection among adult females [49]. In Germany, Kleemann et al. (2022) observed a significant decrease in hospitalizations for non-melanoma skin cancer among females, but a significant increase in cases of malignant melanoma in populations without age restriction [48].

Various pathologies of different groups

Two studies addressed various conditions spanning multiple diagnostic categories. In pediatric populations, Kim et al. (2022) in Korea analyzed multiple diagnostic categories and found a significant increase in female emergency department visits for infections, mental health issues, and injuries (56). In Italy, Cozzi et al. [2022] reported an increased percentage of female hospitalizations, although without statistical significance (55).

Self-injury

Canada, Saxena et al. (2023) evaluated intentional injuries—including self-injury, violence, aggression, and child maltreatment—in individuals aged ≥ 5 years and found a significant increase in self-injury among males aged 5–10 years [52].

Seng et al. (2023) in the United States studied self-injury in adults and reported an increase among females during the pandemic, though not statistically significant [53]. In.

Sexual abuse

Wong et al. (2022) analyzed emergency department visits for sexual abuse among children aged < 18 years in China. They reported a significant increase in incidence among females during the pandemic, particularly in the 12–17 age group [54].

Suicide attempts

In Japan, Otani et al. (2023) studied individuals aged 13–90 years and reported a decline in the proportion of females who attempted suicide during the pandemic; however, the difference was not statistically significant [51].

Falls

Arslan et al. (2023) examined fall-related emergencies in Turkey among individuals aged ≥ 65 years and found no significant sex-based differences between periods [50].

Symptoms, signs, and abnormal clinical and laboratory findings not elsewhere classified (epistaxis)

In Poland and Romania, Vlaescu et al. (2023) found no statistically significant differences in adult hospitalizations for epistaxis between periods or sexes [59].

Skin and subcutaneous tissue diseases

Temel et al. (2023) in Turkey assessed dermatologic conditions in adults and reported a slight increase in female hospitalizations during the pandemic, though the change was not statistically significant [57].

Diseases of the eye and its annexes

Yilmaz et al. (2022) in Turkey investigated ophthalmic emergencies across all age groups and found a significant reduction in the proportion of females seen during the pandemic, especially among those aged 0–16, 17–40, and 41–64 years. No significant difference was observed among individuals aged ≥ 65 years [58].

Discussion

The WHO advocates for data disaggregation to identify key patterns of health disparities and has urged countries to implement routine health inequality monitoring, including disaggregated data collection [61]. Similarly, Llobera Ribera et al. (2024) and Herten-Crabb et al. (2025) emphasize the inequities experienced particularly by females during the pandemic and the notable absence of gender-based analysis [62, 63]. This systematic review aimed to examine the interaction between patient gender and the timing of hospitalization—either pre-pandemic or during the pandemic—across diagnostic groups. A total of 41 articles addressed this interaction, revealing gender-related differences in the pandemic’s impact across various domains, including social roles, abuse, adherence to hygiene and prevention guidelines, healthcare access, and follow-up of individuals with medical conditions. However, most studies only compared gender differences in hospital admission rates and did not further disaggregate data by gender. Only two articles reported in-hospital mortality rates disaggregated by sex [36, 53], and just one article provided length of hospital stay disaggregated by sex [36]. Beyond differences in hospitalization rates for specific conditions, the reviewed articles highlighted several gender-related health phenomena that emerged under lockdown and other exceptional public health measures. These findings warrant attention from public health systems when caring for populations during lockdown periods.

Increase in gender-based violence and child abuse

The present systematic review identified an increase in gender-based violence and child abuse during the lockdown period [27, 30]. International evidence suggests that lockdown measures may have acted as an amplifier of pre-existing risks of gender-based violence and abuse [64–67]. Factors that may underlie these findings include the disruption of formal help-seeking and detection pathways, as well as increased forced cohabitation with the perpetrator, together with contextual factors — such as mobility restrictions, loss of autonomy, and social isolation — that amplified vulnerability and reduced opportunities to access support [64, 68–71].

Non-clinical indicators, such as the increase in calls to helplines [65, 66] or the greater use of digital channels among young people [67], point to a shift toward alternative forms of help-seeking when in-person pathways were blocked or perceived as unsafe. This transition toward remote channels reflects the need to create safe spaces outside the healthcare system during lockdown and aligns with initiatives implemented in other countries — such as discreet protocols in pharmacies or the expansion of telephone helplines in Italy [64] — suggesting that systems can maintain support pathways even during periods of service disruption.

In this context, some episodes identified in the studies included in this review may have been masked by other reasons for admission, as suggested by patterns observed in cases of maxillofacial fractures [27]. The contrast between the increase in judicial and social indicators of violence [72] and the low detection rates within the healthcare system points to a possible ‘diagnostic displacement’. This phenomenon highlights how the pandemic may have altered entry points to the system, hindering the clinical identification of gender-based and child violence, even when the actual incidence was increasing.

Mental health of children and adolescents

The patterns observed in the articles included in this review within the field of child and adolescent mental health during the pandemic — namely, an overall decline in psychiatric hospitalizations [31, 60] coexisting with an increase in self-harming behaviors in specific subgroups, such as children aged 5 to 10 years [52] — may be attributable to secondary factors arising from lockdown measures. These include social isolation, disruption of school routines, and reduced contact with support networks, which may influence both the expression of distress and the perceived need for care, as well as the ability to access resources. In this context, the increase observed in this review in hospital admissions for eating disorders among adolescent females [60] may be related to prolonged stress, disrupted routines, and a perceived loss of control over the environment.

The lower number of admissions observed among males in several of the studies analyzed in this systematic review [31, 60] may be explained by gender differences in the recognition of psychological distress and in willingness to seek help, characterized by greater awareness and lower self-perceived stigma among females [73]. In addition, the literature indicates that young males are less likely to seek professional help for mental health problems [74] and are less familiar with support services compared with adolescent females [75], which may account for the underutilization of resources in a context where usual access pathways were restricted.

In summary, these observations underscore the importance of strengthening digital infrastructures and accessible support channels, as proposed by Timmis and Brüssow (2020) [76], to ensure continuity of care and safe consultation pathways even during disruptions to in-person services.

Behavioral and healthcare-related factors influencing gender inequalities in adults

In one study included in this review, an increase in hospital admissions for alcoholic liver disease among women was observed in the United States [35]. This finding aligns with reports documenting higher alcohol consumption among adult women during the same period [77, 78]. During the pandemic, changes in health-related behaviors were influenced by factors such as prolonged stress, domestic overload, remote work, and economic instability. These circumstances act as contextual factors that may have amplified pre-existing vulnerabilities, particularly among those simultaneously assuming both work and caregiving responsibilities [77, 78].

Delayed help-seeking among males, commonly described in the literature, may have contributed to the observed increases in admissions and mortality from gastrointestinal bleeding observed in this systematic review [36]. Conversely, the coexistence of longer hospital stays in women, but with lower costs, observed in the same study [36], suggests that care pathways and service intensity differed by gender [79]. Gender differences in the willingness to seek care may have influenced the observed admission patterns [80]. These factors illustrate how healthcare pathways can be shaped by determinants such as differences in perceived severity, availability to access the healthcare system, or consultation thresholds during lockdown.

In this systematic review, declines were observed in cardiovascular interventions, as well as in admissions for cerebral venous sinus thrombosis in women [42, 43] and for non-melanoma skin cancer [48]. These patterns are consistent with limited access to non-urgent care, mobility restrictions, or differences in admission prioritization during the health emergency [81]. Conversely, in this systematic review, increases in hospitalizations for lung cancer among women were observed [49] which may be related to diagnostic delays resulting from reduced availability of testing [82].

In diabetes mellitus, our results show the combination of lower admission rates and higher in-hospital mortality among women [47] which can be interpreted in light of these healthcare determinants: delayed access, more restrictive admission thresholds, and resource limitations during periods of healthcare system saturation [83]. Although epidemiological factors, such as the higher prevalence of type 1 diabetes or obesity in males, explain part of the differences identified in our results [47], healthcare restrictions stemming from system saturation likely played a particularly relevant role in the pattern observed among women during the pandemic.

In summary, changes in health behaviors, help-seeking, and clinical access/prioritization illustrate how the pandemic may have interacted with pre-existing gender inequalities, affecting both individual behaviors and the actual opportunities to use healthcare services. Patterns observed in specific pathologies should not be interpreted in isolation but rather as distinct expressions of multiple shared mechanisms.

Refined segregation of variables interacting with gender to detect inequities

Despite the recommendations by Peters et al. (2023), who emphasize not only the need to disaggregate health data by sex and gender but also to compare them systematically to understand their effects on health outcomes [16] our review indicates limited implementation of this approach.

In this systematic review, this phenomenon was observed in studies on mental disorders in adults, where high diagnostic heterogeneity [32–34] initially made it difficult to detect sex-based inequalities. However, a more refined stratification revealed patterns that did not emerge in the aggregated analysis. In particular, Fluck et al. (2023), using stratification by comorbidity (CCI) and specific diagnoses, identified higher admissions among males with CCI ≥ 2 and among females with self-poisoning [32]. These findings highlight the need to analyze impacts by sex within defined clinical subgroups, as aggregated analyses may obscure significant disparities.

The increase in admissions for mental disorders during the pandemic appears to stem from multiple factors: reduced access to primary care and outpatient mental health services, social isolation, financial hardship, unresolved grief, and potentially increased use of psychotropic medication [32]. These factors may exert differentiated effects based on sex, influenced by structural determinants such as females’ more frequent caregiving roles, males’ lower likelihood of seeking psychological help, and varying exposure to stressors like job loss or domestic violence.

Similarly, in our results, the identification of an increase in self-harming behaviors among children aged 5 to 10 years [52] was made possible through stratification by defined age groups, underscoring the value of disaggregation in revealing hidden inequalities in aggregated analyses.

Another example of this systematic review comes from the study by Duhan et al. (2023), who analyzed the interaction between race and gender and observed during the pandemic that African American women had the highest incidence of ischemic events [41]. Authors of another study emphasize that, while part of this difference may be linked to biological susceptibility, the structural role of racism in shaping social determinants of health contributes to higher cardiovascular morbidity and mortality [84]. These findings illustrate that inequalities cannot be explained solely by sex, but rather by the intersection of multiple axes of inequity.

Some benefits of containment measures by gender

Containment measures implemented during the pandemic substantially reduced exposure to everyday activities associated with risk — such as mobility, work-related travel, sports, and social interaction—and altered people’s engagement with the healthcare system [21, 24, 27]. The declines observed in various causes of hospital admission, with effects differing by gender, may be explained by these changes as contributing factors.

In this context, our results showed a decrease in traffic- and work-related injuries [21, 24, 27] reflecting a generalized reduction in exposure, reinforced by a shift toward community-based care [26] and reduced hospital use due to fear of COVID-19 infection [20]. Additionally, the results showed in this systematic review suggest that school closures, reduced social interaction, and improved hygiene practices may have contributed to declines in common infections, such as urinary tract infections in girls [45] or infectious conjunctivitis in children, likely more pronounced among girls. Similarly, our results showed the reduction in ophthalmologic admissions among women under 65 can be understood as a reflection of these exposure changes, whereas in young adults, ocular trauma remained more frequent among males due to higher participation in sports and risk-prone activities [58].

Finally, some observed declines in this systematic review appear to have been influenced by lockdown-specific behavioral changes, particularly among males. Reduced opportunities for social alcohol consumption [35, 92] combined with the tendency described in other studies to minimize symptoms and avoid healthcare in crisis situations [85–87], may have acted jointly as modulatory factors in the decrease in hospitalizations for alcoholic liver disease among males during the pandemic.

Overall, these patterns demonstrate that containment measures not only reduced exposure to risks but did so in a gender-differentiated manner, highlighting the need to consider how prior activities, social roles, and healthcare behaviors influence vulnerability during periods of restriction.

Absence of registration of intersex and transgender individuals

Sex is typically recorded in a binary format in health records, which does not reflect biological reality. Intersex individuals, who comprise approximately 1% of the population, remain understudied, particularly regarding how their hormone profiles and chromosomal variations may confer susceptibility or resilience to diseases and pharmacologic therapies. Additionally, limited research exists on how biological sex and gender roles intersect in the morbidity and mortality of transgender individuals. This gap arises largely because databases rarely document both gender identity and biological sex simultaneously [88–90].

For example, Latasa et al. (2020), in a study conducted in Spain, were unable to identify transgender individuals through the Minimum Basic Data Sets (MBDS) due to the absence of recorded gender identity. Instead, they relied on ICD-9 codes for sexual and gender identity disorders. The authors acknowledged that this approach excluded transgender individuals lacking a formal transsexuality diagnosis, but no alternative method of identification existed in the available health data. This limitation significantly hampers both research and public health planning [91].

Strengths and limitations

This systematic review, to our knowledge the first to examine the gender-based impact of the COVID-19 pandemic on hospitalizations, offers strong evidence that the health crisis affected males and females differently. The included studies indicate that gender roles, inequitable healthcare access, and lockdown-related restrictions significantly influenced hospitalization trends. These findings shed light on pre-existing structural inequities and offer essential insight for developing more equitable health policies during future public health emergencies.

Nevertheless, the review has limitations that must be considered. Diagnostic classifications, age group categorizations, definitions of the pandemic period, and geographical variability among the studies limit the identification of consistent patterns. Moreover, most studies did not adjust analyses for comorbidities or social determinants of health.

Similarly, this systematic review does not allow for establishing causal relationships between sex/gender and the observed hospitalization patterns. Based on observational and heterogeneous studies, our aim was to describe and identify gender inequalities across different diagnostic groups and periods of the pandemic, without attempting to determine what portion of these differences is attributable to biological mechanisms, behavioral factors, access barriers, or organizational aspects of the healthcare system. The identified patterns generate relevant hypotheses about the potential pathways through which sex/gender may influence hospital care; however, verification of these mechanisms requires experimental, quasi-experimental, or longitudinal studies with adequate control of confounding factors and precise definitions of sex and gender. Therefore, future research with more robust designs is needed to analyze more clearly the underlying causality of the observed inequalities.

Conclusions

This review not only highlights differences in hospital admissions across diagnostic categories but also underscores the health and social inequities intensified by lockdown conditions. Overall, females experienced higher hospitalization or in-hospital mortality for several conditions, as well as more frequent emergency presentations related to infections, mental health conditions, and injuries, whereas males showed increased admissions for gastrointestinal hemorrhage [28, 35, 36, 47–49, 56].

In line with the conceptual premises outlined in the introductory literature and international recommendations, these patterns are consistent with the interaction of biological factors and gender‑related social roles, health behaviors, and access to care, rather than biological sex alone. The observed shifts in admissions likely reflect, at least in part, the combined effects of altered care‑seeking behaviors, social isolation, increased exposure to gender‑based violence, and health system reorganization under lockdown conditions. Several studies also stress the need for data disaggregation by factors that interact with gender, such as age, comorbidity burden, race, and diagnostic specificity, to more precisely identify disparities, as advocated by the WHO [1]. From a public health standpoint, the results of this systematic review highlight the need for equity‑oriented monitoring and for preparedness strategies that protect access to care and support targeted telehealth and outreach interventions during future public health emergencies.

In terms of generalizability, some of the observed patterns observed in this systematic review — such as the reduction in admissions for trauma or certain infections linked to decreased exposure during lockdown—could be extrapolated to other contexts where similar measures were implemented. However, other findings, particularly those related to gender-specific patterns in cancer, diabetes, mental health, or alcoholic liver disease, should be interpreted in light of the epidemiological, organizational, and sociocultural characteristics of the healthcare systems included in the analyzed studies. Therefore, not all identified inequalities can be generalized uniformly, highlighting the need for context-sensitive analytical frameworks and further research to clarify variability across countries and populations.

Authors’ contributions

NV: Data curation, Investigation, Methodology, Writing - Original Draft. HA Conceptualization, Project administration, Supervision, Writing - Review & Editing. IZ Investigation, Methodology, Writing - Original Draft. MK: Investigation, Methodology, Writing - Original Draft. AL: Conceptualization, Project administration, Supervision, Writing - Review & Editing.

Funding

Grant PID2024-156774OA-I00 funded byMICIU/AEI/10.13039/501100011033 and by "ERDF/EU".

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.