Menopause age and hypercholesterolemia comorbidities: a region-wide retrospective cohort study in Andalusia, Spain (2016–2022) Authors: Alberto Esteban-Medina, Víctor de la Oliva, Patricia Fernández del Valle, Ana Sánchez, M Belen Susin, Carlos Loucera, Guillermo Antiñolo, Joaquín Dopazo Published: 2026 ABSTRACT Abstract AbstractObjectivesTo quantify sex- and age-related differences in hypercholesterolaemia diagnosis and associated comorbidities around the menopausal transition, using a population-based real-world dataset.DesignRetrospective, multicentre, non-interventional observational cohort study.SettingRegion-wide public healthcare system data (primary and secondary care) from Andalusia (Spain), 2016–2022.ParticipantsAll adult patients meeting inclusion criteria with a recorded diagnosis of hypercholesterolaemia between 1 January 2016 and 31 December 2022 (n=557 034; 227 834 men and 329 200 women).InterventionsNone.Primary and secondary outcome measuresPrimary outcomes were age- and sex-stratified patterns of hypercholesterolaemia diagnosis and comorbidity burden before and after age 50 years (proxy for post-menopausal age). Secondary outcomes included comorbidity-specific comparisons between sexes across age strata and trajectory-based analyses (OR trajectories and incidence-ratio summaries).ResultsWomen were diagnosed later than men (mean age 59.1 vs 56.0 years; mean difference 3.1 years, 95% CI 3.03 to 3.17). Hypercholesterolaemia diagnoses in women rose sharply around ages 50–55 and remained higher than in men at older ages. Comorbidity patterns differed by sex across age strata: compared with men, women aged ≥50 years had higher frequencies of osteoporosis (42 255 vs 2623), anxiety disorder (94 916 vs 31 374) and hypertension (147 538 vs 91 532), with statistically significant differences for these comparisons (p<0.001).ConclusionsMenopause age is a pivotal period associated with a shift towards higher hypercholesterolaemia diagnosis rates and a greater burden of specific comorbidities in women. These findings support sex-specific prevention and management strategies, particularly targeting the menopausal transition and early post-menopause.Keywords: EPIDEMIOLOGY, EPIDEMIOLOGIC STUDIES, PUBLIC HEALTH, SEXUAL MEDICINE STRENGTHS AND LIMITATIONS OF THIS STUDY This study used one of the largest, population-based, region-wide health-record linkage datasets, spanning primary and secondary care, enabling comprehensive capture of routine clinical diagnoses. This study used one of the largest, population-based, region-wide health-record linkage datasets, spanning primary and secondary care, enabling comprehensive capture of routine clinical diagnoses. Hypercholesterolaemia and comorbidities were defined using prespecified International Classification of Diseases, version 10 (ICD-10) code lists applied uniformly across the full cohort. Hypercholesterolaemia and comorbidities were defined using prespecified International Classification of Diseases, version 10 (ICD-10) code lists applied uniformly across the full cohort. Eligibility required evidence of regular follow-up within the health system (at least one annual health record during the study period), reducing inclusion of sporadic users. Eligibility required evidence of regular follow-up within the health system (at least one annual health record during the study period), reducing inclusion of sporadic users. Menopause was not directly observed; analyses used age (≥50 years) as a proxy, which could misclassify sometimes menopausal status and attenuate associations. Menopause was not directly observed; analyses used age (≥50 years) as a proxy, which could misclassify sometimes menopausal status and attenuate associations. Reliance on routinely coded diagnoses and limited availability of key clinical covariates (eg, laboratory lipid values, symptom onset and treatment adherence) may introduce information bias and residual confounding. Reliance on routinely coded diagnoses and limited availability of key clinical covariates (eg, laboratory lipid values, symptom onset and treatment adherence) may introduce information bias and residual confounding. Introduction Historically, medicine has relied on pattern recognition largely shaped by male-normative perspectives, which often neglect the diversity of human physiology. This bias has hindered a comprehensive understanding of sex- and gender-based differences in health.1 Recent studies report higher burden of morbidity-driven conditions in females compared with males, pointing to an urgent need for policies to be based on sex-specific and age-specific data.2 Ageing presents distinct challenges for men and women. Actually, women, while generally living longer, experience higher rates of chronic diseases and multimorbidities.3 Hormonal, biological and social factors significantly contribute to these health disparities, yet knowledge gaps remain.4 Fertility is not just a biological process; it is a fundamental aspect of women’s health throughout their lives. It is essential to rediscover, redefine and understand women’s fertility in a broader context, particularly in relation to overall women’s health. Fertility is not limited to reproduction; it is an indicator of the body’s overall functioning and women’s physiology.5 Therefore, it is reasonable to consider that fertility, normal ovarian function and, consequently, the production of oestrogens, the primary female sex hormones, play a central role in women’s physical and psychological well-being. When fertility is lost, the greatest challenge is the decrease in oestrogen, which is the ‘engine’ of women’s health. These hormones are essential for the female body, and their role extends far beyond reproduction. The longitudinal study of hypercholesterolaemia patients before and after the menopause period, when hormonal changes occur in women, constitutes an excellent scenario to understand the disparities in disease progression between sexes. The decline in oestrogen during menopause significantly impacts lipid metabolism, contributing to adverse changes in cholesterol profiles and an increased risk of hypercholesterolaemia.6 Oestrogen’s protective effects on cardiovascular health are well-documented, and its loss during menopause exacerbates vulnerability to conditions like atherosclerotic cardiovascular disease.3 These hormonal shifts highlight the critical link between fertility loss and the progression of hypercholesterolaemia and related comorbidities in women, underscoring the importance of sex-specific approaches to its management and treatment. Understanding the sex-specific differences in its progression, associated comorbidities and management is essential for effective treatment strategies. From childhood, females with familial hypercholesterolaemia exhibit higher total cholesterol and low-density lipoprotein cholesterol (LDL-C) levels compared with males, having a greater cumulative LDL-C burden by age 30.7 Despite this, women are often diagnosed later than men, delaying the initiation of lipid-lowering therapies.8 In primary care settings, women are 22% less likely than men to achieve LDL-C targets within 180 days of starting or adjusting lipid-lowering therapy, regardless of therapy intensity, cardiovascular risk category, age, mental health status or socioeconomic factors.9 Moreover, post-menopausal women experience a significant increase in LDL-C levels, attributed to decreased oestrogen, which adversely affects lipid profiles.10 This elevation in LDL-C heightens the risk of developing comorbid conditions such as hypertension, type 2 diabetes and metabolic syndrome.11 Additionally, women with hypercholesterolaemia are at increased risk for osteoporosis and certain cancers, including breast cancer, due to complex interactions between lipid metabolism and hormonal changes.11 Large-scale real-world data (RWD) can complement clinical trials by characterising disease trajectories and multimorbidity patterns in broad, unselected populations, including groups typically under-represented in trial settings.12 13 However, population-level evidence specifically describing how hypercholesterolaemia diagnosis patterns and comorbidity profiles shift across menopause age remains limited, particularly using trajectory-based approaches that capture age-varying sex differences. This study aimed to (1) describe sex differences in age at diagnosis of hypercholesterolaemia, (2) compare comorbidity burden and selected comorbidities between women and men before and after age 50 years (as a pragmatic proxy for post-menopausal age) and (3) evaluate age-varying sex differences in comorbidity patterns using trajectory-based analyses to identify groups of comorbidities with similar sex-by-age profiles. Ageing presents unique health challenges for both men and women, but their disease trajectories, the onset of comorbidities and their progression substantially differ. RWD can facilitate the investigation of these differences, as traditional clinical trials often fail to capture the diversity of the population or reflect the daily experiences of men and women.12 13 Andalusia is the largest region in Spain and the third largest in Europe, with a population of 8.5 million, comparable to countries like Switzerland and Austria. The Andalusian Health System has been collecting medical data from their used since 2001 in the Population Health Database (BPS, Spanish acronym for Base Poblacional de Salud).14 The findings reveal a remarkable difference between men and women both in the evolution of the disease and the associated comorbidities. These results will be useful for future clinical recommendations and will contribute to improve the management of the patients, ultimately enhancing their quality of life. This research also underscores the transformative potential of RWD15 in refining healthcare practices and guiding decision-making in patient care. Material and methods Study design and setting We conducted a retrospective, multicentre, non-interventional observational cohort study using routinely collected RWD from the public health system in Andalusia (southern Spain). Data source Data were obtained from the Population Health Database (BPS),14 a region-wide repository that integrates routinely collected clinical information from the Andalusian Health System. For this study, BPS provided pseudonymised individual-level data on diagnoses and selected clinical characteristics, enabling the evaluation of comorbidity patterns and their variation by age and sex. BPS integrates electronic medical records and data covering hospital activity including inpatient admissions, hospital emergencies and medical day hospital. It also includes resource use in Primary Care and Specialist Care visits, hospital episodes and pharmacy consumption, which covers the typical outpatient activity. It also incorporates additional clinical information systems such as mental health, laboratory and imaging tests, vaccination registries, renal patient registries and pharmacy data. Study period The study period was 1 January 2016 to 31 December 2022. Participants, index date and follow-up The target population comprised adult (≥18 years) patients with a recorded diagnosis of hypercholesterolaemia in BPS during the study period (International Classification of Diseases, version 10 (ICD-10) codes: E78.0, E78.00, E78.01, E78.2, E78.49). The index date was defined as the first recorded diagnosis of hypercholesterolaemia during the study period. Participants were followed from the index date until death, loss to follow-up or the end of the study period, whichever occurred first. Menopause most commonly occurs between the ages of 45 and 55,16 and studies conducted in Spain place the average age of natural menopause at around 51 years.17 Therefore, this study has used this age as the starting point for menopause. To enhance data completeness and continuity of observation, we required evidence of regular follow-up (at least one annual health record during the study period). Patients receiving palliative care at index (including ICD-10 code Z51.5) were excluded. Variables used Clinical data, along with comorbidities, defined as the occurrence of one or more diseases (in addition to hypercholesterolaemia) of different categories (cardiovascular, dermatological, bone and joint, neurological, renal and urinary, mental health and substance use, liver, ocular, cancers, endocrine and metabolic, gastrointestinal, collagen disease and vasculitis, cerebrovascular, respiratory, congenital and developmental disorders, infectious diseases), obesity, current smoker status and alcohol consumption and their corresponding diagnosis dates, were used (see online supplemental table 1 for a detailed list of the comorbidities used). Individuals with missing variables were discarded. Statistical methods, sample size and secure data management To identify differences between women and men, categorical variables were presented as frequencies and percentages. Continuous variables were summarised using means and SD or medians and IQRs, depending on the data distribution. The X2 or Fisher’s exact test was used to compare categorical variables, while the t-test or Wilcoxon rank-sum test was used to compare continuous variables, depending on their distribution. The Kruskal–Wallis test was used to compare cost differences between more than two groups. All analyses were two-sided, and statistical significance was set at p<0.05. To estimate incidences, the base distribution of population in Andalusia across ages has been obtained from the Spanish National Institute of Statistics.18 Thus, incidences were approximated as the observed diagnosis cases in the BPS (covering over 98% of the Andalusian population14 at each age were divided by the total population in Andalusia at these ages). To further investigate the relationship between the number of diagnoses, age and sex, an analysis of the OR trajectories was conducted. The data were stratified by age and sex, and the number of diagnoses within each stratum was determined. For each age, the odds of diagnosis in women relative to men was calculated. CIs (95%) for each OR were constructed using the normal approximation. The analysis used statistical methods to investigate trends in disease prevalence across sex and age. Each disease trajectory was modelled after the logarithmic ORs, reflecting the relative prevalence of diseases between women and men across the 35–65 age range. K-means clustering was employed to group diseases with similar log OR trajectories over age, with the optimal number of clusters determined using the elbow method.19 The statistical analysis used Python (V.3.10.12) and open-source libraries, including Numpy (V.2.1.1),20 SciPy (V.1.14.1),21 statsmodels (V.0.14.3)22 and TableOne package (V.0.9.1).23 Trajectory clustering was done using scikit-learn (V.1.5),24 while kneed (V.0.8.5)25 was used to compute the optimal number of clustering trajectories. Data visualisations, such as boxplots and bar charts, were generated using Seaborn (V.0.13.2).26 Sample size estimation In conventional studies using hospital databases, the size calculation underlies the need to obtain a portion or subset of individuals with specific characteristics (ie, sample) that is representative of the whole and to be able to respond to the study objectives with adequate power and precision so that the results can be generalised to the whole population. This study uses RWD, and the major difference in this approach is precisely that this study allows the inclusion of all patients who meet the selection criteria, that is, the sample is directly the entire target population of hypercholesterolaemia patients in Andalusia. However, an assessment is made of the sample size required for the study. The prevalence of hypercholesterolaemia in the Spanish adult population has been estimated at 50%.27 With this prevalence, the minimum required sample size would be 1187 for the margin of error or absolute precision of 3% in estimating the prevalence with 95% confidence and considering the potential loss/attrition of 10%. The anticipated 95% CI is (47%, 53%). This minimum sample size is calculated using the ScalaR SP.28 The sample used in this study far exceeds the minimum required sample size. Secure data management The data management circuit was designed to minimise the relative risk as described in the Impact Assessment on Data Protection analysis29 and following the regulation for the use of medical data for research purposes in Andalusia (Joint Resolution 1/2021 of the General Secretariat for Research, Development and Innovation in Health of the Regional Ministry of Health and Families and the Management Directorate of the Andalusian Health Service).30 Briefly, the data corresponding to the study approved by the ethics committee are requested to BPS. There, the data are extracted and pseudonymised by BPS personnel and then transferred to a Secure Processing Environment (SPE),31 specifically designed for the secure analysis of data protected by the General Data Protection Regulation (GDPR), being compliant with the definition of SPE as described in Article 50 of the Resolution of the European Parliament of 24 April 2024 on the proposal for a Regulation of the European Parliament and of the Council on the European Health Data Space (EHDS).32 Patient and public involvement None. Results Cohort of study A total of 557 034 patients diagnosed with hypercholesterolaemia meeting the inclusion criteria between 1 January 2016 and 31 December 2022 were found. This cohort is composed of 227 834 men and 329 200 women, and the average age (and SD) of hypercholesterolaemia diagnosis was 56.0 (13.2) and 59.1 (13.2) years old, respectively. Table 1 summarises the basic data of the cohort studied, including the different comorbidity categories (see online supplemental table 1 for the whole list of comorbidities). As expected, differences between sex and age onset are significant for all the comorbidity types, except for the case of age onset in infectious diseases. Online supplemental table 2 contains the detailed list of comorbidities for each disease category. Table 1Basic data of the cohort studied, along with the comorbidities by categories, with the mean age at diagnosis with the [Q1,Q3] IQR, and the number of individuals diagnosed with this comorbidityComorbidity categoryDiagnosisMenWomenP valueCardiovascular diseasesAge (Q1, Q3)57.0 (49.0, 65.0)59.0 (51.3, 68.0)<0.001N (%)141 374 (62.3)196 303 (59.9)<0.001Dermatological disordersAge (Q1, Q3)52.0 (42.0, 61.0)54.0 (45.0, 62.0)<0.001N (%)12 445 (5.5)16 432 (5.0)<0.001Bone and joint disordersAge (Q1, Q3)59.0 (51.0, 66.0)59.0 (52.0, 66.3)<0.001N (%)89 328 (39.4)205 724 (62.8)<0.001Neurological disordersAge (Q1, Q3)65.0 (54.0, 74.9)70.0 (58.0, 78.0)<0.001N (%)26 159 (11.5)40 214 (12.3)<0.001Renal and urinary disordersAge (Q1, Q3)57.0 (47.0, 68.0)58.0 (48.0, 70.0)<0.001N (%)47 767 (21.0)67 637 (20.6)<0.001Mental health and substance useAge (Q1, Q3)52.0 (43.0, 62.0)54.0 (45.5, 64.0)<0.001N (%)112 541 (49.6)194 227 (59.3)<0.001Liver diseasesAge (Q1, Q3)55.0 (47.0, 64.0)60.0 (53.0, 67.5)<0.001N (%)31 180 (13.7)35 140 (10.7)<0.001Ocular diseasesAge (Q1, Q3)63.0 (54.0, 72.0)65.2 (57.0, 74.0)<0.001N (%)24 819 (10.9)40 048 (12.2)<0.001CancersAge (Q1, Q3)65.0 (57.0, 73.0)64.0 (55.0, 73.0)<0.001N (%)15 374 (6.8)15 345 (4.7)<0.001Endocrine and metabolic disordersAge (Q1, Q3)52.0 (44.0, 61.0)56.0 (48.0, 65.0)<0.001N (%)145 737 (64.2)191 320 (58.4)<0.001Gastrointestinal disordersAge (Q1,Q3)54.0 (45.0, 64.0)58.0 (50.0, 67.0)<0.001N (%)25 635 (11.3)50 344 (15.4)<0.001Collagen disease and vasculitisAge (Q1, Q3)64.0 (53.0, 73.0)61.0 (53.0, 71.0)<0.001N (%)3107 (1.4)11 820 (3.6)<0.001Cerebrovascular disordersAge (Q1, Q3)68.0 (60.5, 76.0)71.0 (62.0, 79.0)<0.001N (%)7022 (3.1)6321 (1.9)<0.001Respiratory diseasesAge (Q1, Q3)58.0 (47.0, 67.0)56.0 (46.0, 66.0)<0.001N (%)42 893 (18.9)56 651 (17.3)<0.001Congenital and developmental disordersAge (Q1, Q3)57.0 (48.0, 66.0)59.0 (47.0, 70.0)<0.001N (%)1484 (0.7)1552 (0.5)<0.001Infectious diseasesAge (Q1, Q3)42.0 (36.0, 49.0)41.0 (35.0, 49.0)0.055N (%)1399 (0.6)590 (0.2)<0.001Total patients226 962327 710Differences between men and women are tested for ages with a Kruskal-Wallis test and for number of affected by a X2 test. Basic data of the cohort studied, along with the comorbidities by categories, with the mean age at diagnosis with the [Q1,Q3] IQR, and the number of individuals diagnosed with this comorbidity Differences between men and women are tested for ages with a Kruskal-Wallis test and for number of affected by a X2 test. Influence of sex in the hypercholesterolaemia diagnosis Figure 1A illustrates the incidence of hypercholesterolaemia diagnoses across age groups, separated by sex (see online supplemental table 3 with the values used in the estimation). Several key patterns emerge from this data that provide insights into the prevalence and progression of this condition. first, an age-related pattern steady increase of hypercholesterolaemia diagnoses with age in both sexes is observed, peaking between 50 and 70 years, followed by a gradual decline in older populations. The bell-shaped curve is consistent with age-related metabolic changes, dietary patterns and increased screening in middle-aged individuals. Diagnoses and comorbidities across ages. (A) Incidence of hypercholesterolaemia, as diagnoses per 100 individuals, in the cohort under study across ages, stratified by sex (men and women). (B) Incidence of comorbidities, as diagnoses per 100 individuals, in the cohort under study across ages, stratified by sex (men and women). However, sex-specific patterns are dominant, with diagnoses in men increasing more rapidly at younger ages (30–50 years), surpassing those of women in early adulthood. The incidence in men peaks slightly earlier (around 50–55 years), suggesting that hypercholesterolaemia may manifest earlier in males due to factors such as lifestyle, diet or inherent biological differences (eg, higher baseline LDL-C levels). Diagnoses in women increase more gradually but surpass those of men around 50–55 years and remain higher in older age groups. After menopause, women show consistently higher rates of hypercholesterolaemia than men, reflecting long-term cardiovascular risk. Beyond the age of 70, the incidence of diagnoses declines for both sexes, likely due to survivorship bias (healthier individuals are overrepresented in older cohorts) or reduced screening frequency. Influence of sex in the comorbidity frequency Figure 1B illustrates the incidence of diagnosed comorbidities across age ranges, stratified by sex (men and women) in the studied cohort of hypercholesterolaemia patients (see online supplemental table 4 with the values used in the estimation). The general trend indicates a clear age-related increase in the number of comorbidities, peaking in middle age and gradually declining in older age groups. There are some notable differences between men and women on menopause and sex-based health disparities. As an overall trend, both sexes exhibit an increase in the number of comorbidities with age. The peak incidence of comorbidities occurs around the age of 50–70 for both men and women. However, there are remarkable sex-specific differences: women consistently display a higher incidence of comorbidities across nearly all age ranges, particularly between the ages of 45 and 70. The difference becomes particularly pronounced around the menopausal transition period (ages 45–55). Sex-related comorbidity pattern in the post-menopause age In order to accurately quantify the comparative occurrence of different comorbidities during the menopause period, the proportion of diagnoses in women versus men was assessed. Figure 2A portraits the ORs of the total occurrence of comorbidities in women versus men, in which a significant trend of increase in women morbidities with respect to men is observed immediately after the 50 years. This global trend is followed to some extent by many of the comorbidities studied. Comorbidity ORs trajectories. (A) ORs and CIs of the total occurrence of comorbidities in women vs men. (B) Adjusted log-OR of the occurrences of comorbidities in women versus men. Two clusters of diseases are represented. When individual OR trajectories for various morbidities are analysed using K-means clustering (see Methods), two primary trends emerge, as depicted in figure 2B. One cluster predominantly includes sex-dependent conditions, encompassing mental, cardiovascular, endocrine, metabolic, bone and joint, neurological and liver disorders. The other group includes sex-independent diseases, some of them age-related, such as cancers (see figure 2B) or intellectual disability, and others related to genetic factors, such as congenital cardiac and circulatory anomaly or infectious, such as HIV. Online supplemental table 5 contains a detailed description of the diseases following both trends depicted in figure 2B. Table 2 lists the most frequent comorbidities (those accounting for more than 1.5% of the patients with one or more diagnoses) observed in the cohort under study (a complete list is available in online supplemental table 6). The data highlight that the prevalence of comorbidities increases with age in both sexes, but this rise is disproportionately higher in women post-menopause and in most cases statistically significant. Mental and behavioural disorders, such as anxiety, other organic mental disorders and mood disorders, consistently exhibit higher prevalence in women, with notable post-menopausal increases in conditions like dementia. Cardiovascular diseases also show marked post-menopausal increases, with heart failure, atrial fibrillation, peripheral arterial disease and ischaemic heart disease highlighting the loss of oestrogen’s protective effects. Bone-related conditions, such as osteoarthritis, spondylosis and osteoporosis, along with metabolic disorders like pure hyperglyceridaemia, obesity or metabolic syndrome further underscore the significant correlation observed between menopause and women’s health. Table 2The most frequent comorbidities in womenComorbidities<50≥50P valueRatio 95% CI %MenWomenMenWomenOsteoarthritis, spondylosis14 24728 40566 032156 067<0.0011.19(1.16 to 1.21)12.90Hypertension39 13838 77691 532147 538<0.0011.63(1.60 to 1.65)12.20Anxiety disorder29 69564 87531 37494 916<0.0011.38(1.36 to 1.41)7.85Hypothyroidism557728 85510 16157 628<0.0011.10(1.06 to 1.14)4.76Diabetes14 35810 84544 87453 588<0.0011.58(1.54 to 1.63)4.43Mixed hyperlipidaemia41 02522 74434 20745 362<0.0012.39(2.34 to 2.44)3.75Osteoporosis3222081262342 255<0.0012.49(2.20 to 2.82)3.49Mood disorder629514 45610 79841 216<0.0011.66(1.60 to 1.72)3.41Obesity13 27815 70122 77236 240<0.0011.35(1.31 to 1.38)3.00Urinary stones14 12818 29621 03034 946<0.0011.28(1.25 to 1.32)2.89Gastro-oesophageal reflux disease (GORD)7384799912 48330 642<0.0012.27(2.18 to 2.35)2.53Tobacco dependence22 16218 43728 91923 832 0.48 0.99(0.97 to 1.02)1.97Asthma992216 714793423 426<0.0011.75(1.69 to 1.82)1.94Other arthropathy5452594313 13323 126<0.0011.62(1.55 to 1.69)1.91Heart failure113871116 41222 915<0.0012.23(2.03 to 2.46)1.89Chronic Obstructive Pulmonary Disease4426392925 49222 087 0.313 0.98(0.93 to 1.02)1.83Hepatic steatosis (Fatty liver)7472401415 30721 695<0.0012.64(2.53 to 2.76)1.79Pure hyperglyceridaemia24 51511 33918 12820 303<0.0012.42(2.35 to 2.50)1.68Fibromyalgia55810 74982320 035<0.0011.26(1.13 to 1.41)1.66Glaucoma19322861942419 129<0.0011.37(1.29 to 1.46)1.58Atrial fibrillation126851415 07318 787<0.0013.07(2.77 to 3.41)1.55Number of patients with at least one comorbidity (second and third columns represent observations in patients younger than 50 years, and fourth and fifth columns in patients 50 years old and older). A X2 test has been conducted to detect significant increases of comorbidity diagnoses in women with respect to men when the older (post-menopausal age) are compared with the younger, and the sixth column reports the p value. The seventh column represents the total incidence ratio woman/man between the respective increase in comorbidities in older with respect to younger, followed by a column with the 95% CI. The last column is the percentage of women with the comorbidity with respect to all women with comorbidities. The most frequent comorbidities in women Number of patients with at least one comorbidity (second and third columns represent observations in patients younger than 50 years, and fourth and fifth columns in patients 50 years old and older). A X2 test has been conducted to detect significant increases of comorbidity diagnoses in women with respect to men when the older (post-menopausal age) are compared with the younger, and the sixth column reports the p value. The seventh column represents the total incidence ratio woman/man between the respective increase in comorbidities in older with respect to younger, followed by a column with the 95% CI. The last column is the percentage of women with the comorbidity with respect to all women with comorbidities. Among cancers, most conditions exhibit a significant age-related increase in both sexes, as anticipated, with few exceptions showing sex-specific differences. Online supplemental table 4 highlights that non-Hodgkin’s lymphoma (p=0.01) and leukaemia (p=0.003) have statistically significant higher prevalence in women compared with men post-menopause. Additionally, marginally significant differences are noted in bladder cancer (p=0.031) and bronchus and lung cancer (p=0.044). Table 3 and the OR plots in the online supplemental figure 1A illustrate the pronounced disparities in comorbidities between men and women in the post-menopausal period. Conditions like gout and other crystal-induced arthropathies (woman-to-man incidence ratio, WMIR=4.65), with a significant OR over 60 years in women (see online supplemental figure 1AD) as well as other organic mental disorders (WMIR=4.02), with a significant OR in women over 70 years (online supplemental figure 1BH), followed by another mental condition, autism spectrum disorder (WMIR=3.44), exhibit the largest disparities, revealing that post-menopausal women face greater relative increases in these conditions compared with men. In general, for mental disorders notable ratios are observed, like schizophrenic disorder (WMIR=2.73), which presents a significant OR in women over 50 years (online supplemental figure 1BQ) and dependence on other substances (WMIR=2.24), with also a significant OR in women over 50 years (online supplemental figure V), indicating that hormonal or neurochemical changes post-menopause may significantly be related with mental health in women and underscoring a heightened vulnerability among women in older age. Among cardiovascular diseases, atrial fibrillation (WMIR=3.07) and ischaemic heart disease (WMIR=2.57) show substantial increases in women, likely due to the loss of oestrogen’s protective effects on cardiac and vascular systems. Both diseases present a significant OR but at older ages, over the 65 years (online supplemental figures 1I and AP, respectively). For metabolic and lipid-related conditions, hepatic steatosis (WMIR=2.74), pure hyperglyceridaemia (WMIR=2.42) and mixed hyperlipidaemia (WMIR=2.39), with a significant OR soon after the menopause period (online supplemental figures 1AH, BM and AW, respectively), highlight the adverse changes on metabolic health in post-menopausal women. Notably, osteoporosis (WMIR=2.49), also with a significant OR soon after the menopause period (online supplemental figure 1BD), reflects the potential relationship between oestrogen loss and bone health, with post-menopausal women disproportionately affected. Overall, these findings emphasise the need for sex-specific prevention and management strategies for women’s health post-menopause. Table 3Diagnoses with most pronounced frequency increase in post-menopausal women with respect to menComorbidities<50≥50P valueRatio 95% CI MenWomenMenWomenGout and other crystal-induced arthropathies660751517 0656180<0.0014.654.23–5.11Other organic mental disorder1435670853116 016<0.0014.023.66–4.42Autism spectrum disorder71253340<0.0013.441.80–6.58Dependence on other substances250275812951343<0.0013.423.06–3.83Atrial fibrillation126851415 07318 787<0.0013.072.77–3.41Schizophrenic disorder2294151012352217<0.0012.732.48–3.00Hepatic steatosis (fatty liver)7472401415 30721 695<0.0012.642.53–2.76Ischaemic heart disease324278920 78313 022<0.0012.572.37–2.79Osteoporosis3222081262342 255<0.0012.492.20–2.82Pure hyperglyceridaemia24 51511 33918 12820 303<0.0012.422.35–2.50Mixed hyperlipidaemia41 02522 74434 20745 362<0.0012.392.34–2.44Dementia107118505813 168<0.0012.361.81–3.07A X2 test has been conducted to detect significant increases of comorbidity diagnoses in women with respect to men when the older (post-menopausal age) are compared with the younger, and the sixth column reports the p value. The seventh column represents the total incidence ratio woman/man between the respective increase in comorbidities in older with respect to younger, followed of a column with the 95% CI. Diagnoses with most pronounced frequency increase in post-menopausal women with respect to men A X2 test has been conducted to detect significant increases of comorbidity diagnoses in women with respect to men when the older (post-menopausal age) are compared with the younger, and the sixth column reports the p value. The seventh column represents the total incidence ratio woman/man between the respective increase in comorbidities in older with respect to younger, followed of a column with the 95% CI. Discussion Principal findings This population-based study using RWD from 5 57 034 patients highlights clinically relevant sex- and age-associated differences in hypercholesterolaemia and comorbidity profiles around menopause age. Consistent with prior evidence, men tended to be diagnosed earlier, whereas women showed a pronounced increase in hypercholesterolaemia and several associated comorbidities after menopause age, supporting the importance of integrating sex-specific perspectives into prevention and management strategies.8 10 These patterns are compatible with the known menopausal transition in lipid metabolism and cardiometabolic risk, in which total cholesterol and LDL-C typically rise during midlife in women beyond what is explained by chronological ageing alone.10 33 Strengths and limitations of this study A principal strength is the use of a large, population-based dataset derived from routine clinical practice, which improves statistical power and captures heterogeneity in disease presentation and multimorbidity that is often not reflected in conventional clinical trials.12 13 This setting also enables evaluation of comorbidity patterns that are highly relevant to clinical workload in primary care and to population-health planning. Several limitations require careful consideration. First, menopause was not directly observed; therefore, the analysis necessarily relies on a pragmatic age-based proxy (ie, menopause age), which can misclassify menopausal status (early/late menopause, surgical menopause and hormone therapy users) and blur the timing of changes. Second, as an observational analysis based on routinely collected health records, causal inference is not possible. The observed sex differences may reflect a combination of biological mechanisms (hormonal transition), healthcare utilisation, diagnostic intensity and treatment patterns rather than purely biological effects. Third, routine coding is susceptible to information bias, including under-recording of conditions managed outside the coded health system, heterogeneity in coding across care settings and temporal variation in coding practices; these issues may be particularly relevant for mental health diagnoses and for risk factors that are inconsistently captured in administrative records. Fourth, residual confounding is likely because key variables were not available or were incompletely captured (eg, lipid laboratory values, menopause timing, hormone therapy exposure, treatment adherence and intensity over time, diet, physical activity and granular socioeconomic measures). Finally, differential detection (women engaging more frequently with healthcare services in midlife) and survivorship/competing-risk effects in older age groups may contribute to the apparent convergence or divergence of some comorbidity patterns and should be considered when interpreting age-stratified findings. Another limitation is that the study population is restricted to individuals with recorded hypercholesterolaemia; therefore, the findings describe sex differences among hypercholesterolaemia patients and may not generalise to the overall population. Selection related to lipid testing, diagnosis and healthcare utilisation could also influence observed patterns. Strengths and weaknesses in relation to other studies Our results are aligned with established literature showing that menopause is associated with adverse lipid changes and increased cardiometabolic vulnerability.10 33 They are also consistent with evidence that women may be diagnosed later and less frequently achieve LDL-C targets in routine care, which could compound risk accumulation over time.8 Differences between our observations and some prior cohort or trial-based reports may be explained by differences in population selection and measurement: clinical trials often enrol healthier, more adherent participants and include systematic lipid testing, whereas routine care reflects symptom- and risk-driven testing and heterogeneous follow-up.12 13 Our findings regarding the higher burden of comorbidities in post-menopausal women (including hypertension, mood disorders and metabolic abnormalities) are in line with prior reports of increased multimorbidity among women later in life and support calls for sex- and age-sensitive approaches to chronic disease management.3 34 Interpretation and implications for clinicians and policymakers From a clinical perspective, these results support the menopausal transition as a practical point to reassess lipid-related risk in women, particularly where diagnosis may be delayed and comorbidity burden is rising.8 10 The co-occurrence of cardiometabolic and mental health disorders observed in post-menopausal women reinforces the need for integrated care pathways that address both physical and psychological health.35 36 This integrated approach is clinically relevant because anxiety and mood disorders can influence adherence, healthcare utilisation and risk-factor control, potentially amplifying long-term cardiovascular risk. At the policy level, our findings support the value of sex- and age-stratified quality indicators for dyslipidaemia management in routine care, including timely identification of hypercholesterolaemia and systematic reassessment of risk at midlife. More broadly, this work illustrates how real-world evidence can complement traditional studies to guide service planning and prevention strategies, while also highlighting that improvement in routinely captured variables (eg, laboratory values, medication exposure and menopause-related fields) would materially increase the actionability of such analyses.12 13 Hormone therapy and personalised prevention Our results are consistent with the broader literature suggesting that hormonal changes occurring in the menopause contribute to midlife shifts in lipid profiles and cardiometabolic risk in women.10 33 Hormone replacement therapy (HRT) has been proposed as a strategy to address menopausal symptoms and to prevent bone loss, with more nuanced evidence regarding cardiometabolic outcomes depending on formulation, route and timing of initiation.37 38 Importantly, major position statements emphasise individualised decision-making and risk stratification by age and time since menopause, rather than using HRT as a universal cardiovascular prevention intervention.37 Within this context, our findings support the need for personalised prevention strategies that integrate lipid management with midlife risk reassessment, multimorbidity screening and patient-centred discussion of benefits and risks. Future research Future studies should incorporate direct measures of menopausal status (age at menopause, surgical menopause and hormone therapy exposure) and longitudinal laboratory lipid profiles to disentangle biological changes from differential testing or coding. Linking medication records would allow evaluation of whether sex differences are driven by treatment initiation, intensity, persistence or adherence. Methodologically, designs that emulate target trials could help clarify how lipid-lowering strategies and/or menopause-related interventions influence subsequent comorbidity trajectories in routine care. Replication in other healthcare systems and stratification by socioeconomic factors would also strengthen generalisability and help identify modifiable drivers of inequity. Supplementary material Acknowledgements We acknowledge the help of Román Villegas and other BPS curators on issues related to the complexity of clinical data management and interpretation. Footnotes Funding: The infrastructure where this work has been carried out has been funded by the Consejería de Salud y Consumo, Junta de Andalucía (IE19_259 FPS) and the Spanish Ministry of Science and Innovation (PRTR-C17.I1). The funder has no role in the study Prepub: Prepublication history and additional supplemental material for this paper are available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2025-101587). Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Not applicable. Ethics approval: This study involves human participants. The Andalusian Biomedical Research Ethics Coordinating Committee approved the study entitled: 'Retrospective observational study to determine the characteristics and treatment patterns in patients with atherosclerotic cardiovascular disease and familial hypercholesterolemia in Andalusia' (28 March 2023, Minutes 03/23). The Andalusian Biomedical Research Ethics Coordinating Committee approved the study (March 28, 2023, Minutes 03/23) and waived informed consent for the secondary use of clinical data for research purposes. Data availability free text: The datasets analysed during the current study are available in the Population Health Database repository (https://www.sspa.juntadeandalucia.es/servicioandaluzdesalud/profesionales/sistemas-de-informacion/base-poblacional-de-salud), subject to controlled access according to the regulation for the use of medical data for research in the Andalusian Health System (https://www.sspa.juntadeandalucia.es/servicioandaluzdesalud/sites/default/files/sincfiles/wsas-media-sas_normativa_mediafile/2021/resolucion_conjunta_acceso_a_datos_investigacion_def_20211201%28F%29.pdf). Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research. Data availability statement Data may be obtained from a third party and are not publicly available. References Supplementary Materials Data Availability Statement Data may be obtained from a third party and are not publicly available.