Association Between Premenstrual Dysphoric Disorder and ADHD/ASD: A Review of Current Evidence

Recent studies have increasingly examined the association between ADHD and PMDD. Dorani et al. [20] reported that in a clinical sample of 209 adult women with ADHD, 45.5% met the PMDD-group criteria, which appears to be higher compared to the 28.7% of participants from the general population observed in a comparison study by Hylan et al. [21]. Although most participants were of reproductive age, a subgroup of peri – and postmenopausal women was also included. PMDD symptoms were assessed retrospectively using thirteen premenstrual dysphoria questions from the Neuropsychiatric Interview Plus version 5.0.0 (the M.I.N.I. Plus) [20].

In a cross-sectional survey by Broughton et al. [22], three groups were identified: a self-reported clinical ADHD diagnosis group, an Adult ADHD Self-Report Scale (ASRS)-based ADHD group, and a non-ADHD reference group. Importantly, the ASRS-based ADHD group included 70.6% of participants who reported a clinical ADHD diagnosis, 49.4% of those who considered themselves to have ADHD, and 14.2% of those who did not consider themselves to have ADHD. PMDD symptoms were assessed using the Premenstrual Symptoms Screening Tool (PSST) which is based on Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition-Text Revision (DSM-IV-TR) criteria. Provisional PMDD was observed in 31.4% of participants reporting a clinical ADHD diagnosis and in 41.1% of those meeting the criteria on the ASRS. In the non-ADHD reference group, provisional PMDD was observed in only 9.8%. These findings correspond to a 3.19-fold higher risk of provisional PMDD among participants with a clinical ADHD diagnosis (95% CI: [2.04, 4.98], p < 0.001) and a 4.17-fold higher risk among those meeting ASRS-based ADHD criteria (95% CI: [2.87, 6.07], p < 0.001). Furthermore, the risk of provisional PMDD was highest among individuals with ADHD and a reported diagnosis of depression/anxiety, both in those with a self-reported clinical ADHD diagnosis and those with ASRS-based ADHD, with statistically significant associations. Similarly, the ASRS-based ADHD group without comorbidities was also significantly more likely to meet provisional PMDD criteria compared with the non-ADHD reference group. Although the results for participants with only a self-reported ADHD diagnosis did not reach statistical significance, they were at twice the risk of provisional PMDD compared with those without an ADHD diagnosis [22].

Evidence from a recent observational case-control study by Lin et al. [23] also suggests a significant association between ADHD and PMDD (p = 0.045). In this study, psychiatric interviews were conducted to diagnose PMDD and adult ADHD according to the DSM-5 criteria. Participants were included only if they had regular menstrual cycles and fulfilled prospective symptom criteria for PMDD. The final sample consisted of 58 women with PMDD and 50 control participants. Among women with PMDD, 27.6% met the diagnostic criteria for ADHD. During the late luteal phase, scores for prospective everyday memory problems and dysfunctional impulsivity did not differ between women with and without ADHD within the PMDD group. However, outside of the late-luteal phase, participants with comorbidity of PMDD and ADHD exhibited significantly greater cognitive and behavioral difficulties compared with those without ADHD. Specifically, they had higher scores for prospective everyday memory problems during both the pre-ovulatory (p = 0.012) and mid-luteal (p = 0.001) phases. In addition, they demonstrated higher levels of dysfunctional impulsivity during the pre-ovulatory (p = 0.006) and mid-luteal (p = 0.004) phases. Interestingly, regardless of ADHD comorbidity, all women with PMDD exhibited greater prospective everyday memory problems and difficulties maintaining focused attention from the pre-ovulatory phase to the late luteal phase compared with healthy controls [23].

Another cross-sectional study by Tsuji et al. [24] involving 2000 full-time female employees in Japan found that ADHD traits, assessed using Part A of the ASRS-v1.1, were present in 8.6% of participants and were strongly associated with PMDD. PMDD was evaluated using a scale based on the research criteria described in the DSM-IV-TR. Overall, the prevalence of PMDD in the sample was 5.1%, and approximately 20.3% of women with ADHD traits met the criteria for PMDD. Women with ADHD traits had a significantly higher likelihood of PMDD compared with those without such traits (adjusted OR = 6.49, 95 % CI: 4.10–10.2, p < 0.001) [24].

Additionally, a retrospective study by Kondo et al. [25] involving 290 adolescent females, diagnosed with ASD, ADHD, or both, suggests that ADHD is associated with greater premenstrual symptom severity. The study used a standardized PMDD assessment scale with self-reported symptoms during the two weeks before menstruation. Patients were classified into three groups: PMDD, moderate premenstrual syndrome (PMS), and none/mild PMS. For analysis, the authors mainly used moderate-or-higher PMS as the dependent variable in logistic regression. Individuals with ADHD had higher odds of moderate-to-severe PMS (OR = 2.43, 95% CI:1.01–5.84, p = 0.047) compared with those without ADHD. However, the authors state that the finding did not remain significant after Bonferroni correction (α = 0.017). Additionally, depressive symptom severity, assessed using the Quick Inventory of Depressive Symptomatology (QIDS), remained strongly associated with PMS severity in women with ADHD (OR = 1.16, 95% CI: 1.08–1.25, p < 0.001) [25].

Evidence on the association between ASD and PMDD remains limited and inconsistent. Early findings from a small prospective, observer-rated study by Obaydi & Puri [26] showed a higher prevalence of late luteal phase dysphoric disorder in autistic women with intellectual disabilities compared to non-autistic participants with intellectual disabilities. All diagnoses were based on DSM-IV criteria. In this study, 92% of autistic participants met the observer-based diagnostic criteria, compared with 11% of matched controls [26].

However, several later studies did not confirm an increased prevalence of PMDD among women with ASD [19, 25]. Groenman et al. [19] found no significant difference in lifetime PMDD prevalence, assessed using the MINI-Plus, between intellectually capable autistic and non-autistic women. Furthermore, Kondo et al. [25] did not identify ASD as a significant predictor of moderate-to-severe PMS. Nevertheless, depressive symptom severity, assessed using the QIDS, demonstrated a strong association with PMS severity also in the ASD-only group (OR = 1.14, 95% CI: 1.07–1.21, p < 0.001) [25]. In addition, in the study by Tsuji et al. [24], participants with autistic traits, assessed by Japanese version of Autism Spectrum Quotient (AQ), had a lower prevalence of PMDD compared to participants with ADHD traits.

In the study by Kondo et al. [25] comorbidity between ASD and ADHD was not identified as a significant risk factor for moderate-to-severe PMS (OR = 2.39, 95% CI: 0.80–7.12, p = 0.12). Nevertheless, depressive symptom severity, assessed using the QIDS, remained a strong predictor of PMS severity (p = 0.006, OR = 1.10, 95% CI: 1.03–1.18) [25]. In contrast, Tsuji et al. [24] reported a significant association between comorbid autistic and ADHD traits and PMDD (adjusted OR = 5.56, 95% CI: 2.14–12.8, p < 0.001) [24].

Growing evidence suggests a potential overlap between ADHD and PMDD, with several studies reporting greater PMDD symptom severity and the possibility of increased PMDD prevalence among women with ADHD or elevated ADHD traits [20, 22,23,24,25]. The association appears to be particularly pronounced in individuals with comorbid depression and/or anxiety, which may indicate that a greater affective burden potentiates menstrual-related symptom expression [22]. However, it should be noted that in the study by Broughton et al. [22], depression and anxiety were assessed based on participants’ responses to whether a clinician had ever diagnosed them with these conditions, rather than through a standardized clinical assessment. Nevertheless, these findings align with Kondo et al. [25], who identified depressive symptoms, assessed by the QIDS, as a strong correlate of PMDD symptom severity. Importantly, this association was observed not only in women with ADHD and PMDD, but also in those with AuDHD and PMDD, as well as in the ASD and PMDD group [25]. By contrast, it remains unclear whether the prevalence of PMDD is increased among individuals with ASD. Although Obaydi and Puri [26] reported a high prevalence of late luteal dysphoric symptoms in autistic women, these findings should not be taken as representative of the broader autistic population, as the study was based on a very small sample, included only individuals with intellectual disability, and applied DSM-IV criteria for autistic disorder rather than the contemporary concept of autism spectrum disorder. In contrast, the more recent study by Groenman et al. [19] did not confirm a significantly increased prevalence of PMDD in intellectually capable autistic adults. Interestingly, that study identified greater menopausal complaints and links between menopausal symptoms, depression, and autistic traits, suggesting that hormonal sensitivity in ASD may be more relevant to the menopausal transition than to PMDD, or may characterize only specific autistic subgroups [19]. Subsequent studies likewise did not confirm an increased prevalence of PMDD among women with autistic traits or ASD [24–25]. While Tsuji et al. [24] observed a significant association between comorbid autistic and ADHD traits and PMDD, Kondo et al. [25] found the link between AuDHD and PMDD to be insignificant. However, the lack of statistical significance in this study may be explained by the relatively small sample size (n = 11) of the comorbid subgroup [25]. Nevertheless, it is essential to acknowledge that interpreting results related to ASD and AuDHD can be complicated by camouflaging (masking), which may reduce apparent identification and diagnosis of autistic individuals [27]. As a result, people with ASD or AuDHD may not be assigned to appropriate study groups. These factors may partly contribute to the inconsistent findings regarding ASD/AuDHD and PMDD in the current literature.

The more consistent association observed between PMDD and ADHD, compared with ASD, could be cautiously interpreted considering the mechanistic models outlined in the introduction. Within a multifactorial framework, PMDD is best described as a centrally mediated disorder in which abnormal brain sensitivity to normal ovarian hormone fluctuations, particularly impaired ALLOmodulated GABA-A receptor function, drives cyclical affective symptoms and heightened physiological stress reactivity [8]. Although altered sensitivity to ALLO has been more clearly implicated in PMDD [8], related neurosteroid sensitivity mechanisms have also been discussed by Wynchank et al. [12] in the broader context of menstrual cycle–related symptom fluctuation in women with ADHD, raising the possibility of a partially shared mechanistic pathway. Although these interpretations remain speculative, they provide a plausible pathophysiological framework that warrants further investigation.

Emerging evidence consistently indicates that PMDD might be particularly common in women with ADHD [20, 22,23,24,25]. Given this significant co-occurrence, routine screening for PMDD is likely warranted in this population [22, 25]. Conversely, Lin et al. [23] suggested that clinicians should consider ADHD comorbidity when evaluating symptoms of inattention and impulsivity in patients presenting with PMDD. However, an important clinical and diagnostic issue is the distinction between PMDD and premenstrual exacerbation (PME) of pre-existing ADHD symptoms. This differentiation carries significant treatment implications: PMDD in the absence of psychiatric comorbidities is reported to be best treated with selective serotonin reuptake inhibitors (SSRIs) and low-dose oral estroprogestins [30], whereas individuals with ADHD may instead require cycle-informed stimulant adjustments, including dosage increases during the premenstrual phase [15]. To improve diagnostic accuracy, studies must evaluate the symptoms outside the late luteal phase. The persistence of deficits across different cycle phases may suggest underlying ADHD. This is supported by Lin et al. [23], who reported that women with comorbid ADHD and PMDD exhibited significantly greater dysfunctional impulsivity and memory impairment during both the pre-ovulatory and mid-luteal phases compared with women with PMDD only. Nevertheless, it is important to remember that an ADHD diagnosis necessitates evidence of symptoms prior to age 12, and their consistent presence across multiple settings [2], rather than manifesting solely in a cyclical, phase-dependent manner.

On the other hand, since it is reported that women with ADHD experience the exacerbation of ADHD symptoms during the luteal phase [12–13, 15–16], cyclical exacerbation of functioning during this phase may represent both ADHD-related PME and PMDD symptoms. However, the differentiation may be especially challenging because many symptoms common in PMDD, including reduced concentration, irritability, affective lability, a sense of being overwhelmed or ‘out of control’, anxiety, changes in appetite, sleep problems, and cognitive impairment [1–2, 10–11], overlap with core and associated features of ADHD such as attention dysregulation, impulsivity, cognitive impairment, emotional dysregulation, low frustration tolerance, and common sleep problems [12, 28–29]. To better clarify the distinctions between PMDD and ADHD-related PME in ADHD population, future studies should employ a three-arm comparison including menstruating individuals with comorbid ADHD and PMDD, those with ADHD only, and those with PMDD only. Furthermore, future research should implement longitudinal monitoring of executive deficits and fluctuations in ADHD trait control across the entire menstrual cycle, as exacerbation of ADHD symptoms was reported to be present not only during the premenstrual (late luteal) phase [12–13, 15], but also during the mid-luteal [12, 14] and the periovulatory [13] phases, with one study suggesting the greatest increase in ADHD symptoms during the early luteal, early follicular, and post-ovulatory phases [16]. Given the profound symptomatic overlap, assessment protocols restricted to the late luteal phase may lack the sensitivity required to distinguish a cyclically amplified baseline from a discrete secondary pathology in ADHD population. Furthermore, there remains a critical evidence gap regarding the optimal pharmacological management of comorbid ADHD and PMDD. Future research should prioritize comparative efficacy trials to evaluate the relative benefits of stimulants, SSRIs, and oral contraceptives, either as monotherapies or in combination, in mitigating symptom exacerbation within this specific population.

While clinical evidence regarding the comorbidity of PMDD and ASD is currently insufficient to mandate routine screening, there remains a need for research to establish evidence-based therapeutic recommendations for patients with comorbid ASD and PMDD. As recent clinical guidelines caution against the use of SSRIs as a first-line pharmacological treatment for depression in autistic individuals, suggesting an elevated risk of behavioral side effects [31], it might be crucial for future research to investigate the comparative efficacy of SSRIs, non-SSRI antidepressants, and oral contraceptives to determine the most effective and safest management strategies for PMDD in this population.

Firstly, the majority of the studies included in this review were cross-sectional, which precludes causal inference. Cross-sectional designs also limit the ability to capture temporal fluctuations in conditions characterized by cyclical symptom variation and distinct phases of expression [24]. To better understand causal mechanisms and long-term mental health effects, future studies should adopt longitudinal frameworks.

Furthermore, most of the studies are limited by self-reported, questionnaire-based and retrospective assessment of PMDD symptoms rather than prospective DSM-5-based confirmation of PMDD across at least two menstrual cycles. Importantly, self-reported and retrospective measurement of PMDD symptoms introduces recall bias and possible symptom over – or underestimation.

Another important limitation is that the ADHD and ASD study groups did not consist exclusively of individuals diagnosed according to DSM-5 criteria. Instead, most of the studies relied on earlier diagnostic frameworks, screening tools, trait-based measures, or self-reported diagnoses. Methodological heterogeneity in ASD and ADHD assessment frameworks may influence prevalence estimates and limit comparability across studies.

Additional limitations included small sample sizes and insufficient control for potentially relevant confounders, such as medication use, lifestyle factors, menstrual cycle regularity, psychiatric comorbidities, environmental influences, and selection bias.

A further limitation is that the available literature primarily concerns cisgender women. Nevertheless, PMDD is relevant more broadly to menstruating individuals, including TGD people [3]. The lack of data in these populations represents an important gap in the literature.