This systematic review was informed by Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and used Covidence, a primary screening and data extraction tool, to extract and screen data (45). Databases MedLine, PsycInfo, Embase, and Web of Science were searched using the following keywords: (depression OR depress* OR MDD OR bipolar* OR bipolar disorder OR ADHD* OR attention deficit hyperactivity disorder OR autism* OR ASD OR anxiety* OR GAD OR mania OR schizo* OR obsessive compulsive disorder OR OCD OR posttraumatic stress disorder OR PTSD) AND (probiotic* OR prebiotic* OR psychobiotics* OR fecal microbiota transplant* OR FMT OR stool transplant* OR bacteriotherapy OR microbe therapy OR microbe transfer OR microbial therap*) AND (follow-up OR long-term OR longitudinal*). The search was conducted from inception up to May 28^th, 2025.

Eligibility criteria for article inclusion is as follows (1): Articles must be written in English and published by a peer-reviewed journal (2); Population: Subjects in the study must be assessed for symptoms of a psychiatric disorder (3); Intervention: Subjects must have undergone a form of microbe transfer and/or received a microbial therapeutic product; (4) Outcome: Subjects must have a follow-up visit that includes clinical assessment conducted at a time point of 12-weeks or longer; (5) Study Design: Restricted to any type of clinical (e.g., RCT) and pre-clinical study. Exclusion criteria was any study that does not meet inclusion criteria.

The initial literature search yielded 4175 studies, of which 1274 duplicates were removed, resulting in 2901 studies to be screened. In total, 70 studies passed abstract screening and underwent full-text review, from which 60 were excluded for either: no assessment of psychiatric symptoms, no follow-up assessment, abstract only, follow-up assessment less than 12 weeks post-treatment, or other type of intervention. This review will include the details of the remaining 10 articles that met eligibility criteria. Please refer to Figure 1 for a detailed description of the above PRISMA screening process. The 10 eligible studies were assessed for quality using the Risk of Bias tool. A detailed summary of the quality assessment can be found in Table 1.

The characteristics and detailed summary of each included study is displayed in Table 2. A total of ten clinical studies with human samples were included in this review, as there were no pre-clinical studies that matched eligibility criteria. The majority of studies in this review primarily explored the effects of microbial therapeutics for GI-related health conditions, with additional outcome measures assessing psychiatric well-being. This included any population that has either assessed for psychiatric disorders (e.g., ASD) or severity of psychiatric manifestations (e.g., symptoms of anxiety). Given the novelty of the field, few studies have assessed the use of microbial therapeutics within a psychiatric setting, let alone examining the associated long-term effects and potential relapse rates.

Seven of the ten included studies in this review investigated physical illnesses as a primary indication and psychiatric symptoms as a secondary outcome (46–52). The studies were comprised of the following primary indications: IBS (n = 3), refractory IBS (n = 1), non-constipated IBS (n = 1), obesity and insulin resistance (n = 1), and recurrent abdominal pain with non-infectious diarrhea (n = 1). The majority of studies involved the use of FMT as the gut-repopulation method from a healthy donor and assessed mood using psychiatric measures. Of the physical indication studies, six studies assessed for depressive symptomology using either the Beck Depression Inventory (BDI), Hamilton Depression Rating Scale (HAM-D), or Montgomery-Asberg Depression Rating Scale (MADRS). The majority of studies reported a sustained improvement in depressive symptoms at least 3-months (n = 4) post-FMT (47, 50–52). Interestingly, Huang et al. (2019) reported a sustained improvement in symptoms at 6-months post-FMT and Xie et al. (2019) found the effect lasting up to 18-months post-FMT, in a single case study (47, 50), while others reported no sustained long-term improvement in depressive symptoms (n = 2) (46, 49).

Again, of the physical indication studies, five assessed for symptoms of anxiety using either the Beck Anxiety Inventory (BAI) or Hamilton Anxiety Rating Scale (HAM-A) and reflected similar mixed findings. Some studies found a sustained improvement in symptoms of anxiety at 3-months post-FMT (n = 2) (47, 51), while the remaining studies did not (46, 49, 52). Some of the physical indication studies (n = 3) also utilized the IBS Quality of Life (IBS-QoL), a questionnaire designed to assess overall quality of life for individuals dealing with IBS, with one study using the IBS-QoL as its sole psychiatric measure. Lahtinen et al. (2020) found no significant effect in IBS-QoL scores across any timepoints, while Lin et al. (2021) found a significant improvement at 3-months post-FMT, and Johnsen et al. (2019) found a clinical improvement in scores at 6-months post-FMT (46, 48, 51).

The remaining psychiatric-focused studies (n = 3) consisted of two case studies (53, 54) and a randomized study (55). As previously mentioned, both case studies have been included due to the novelty of this research area and the lack of available clinical literature. Huang et al. (2022) investigated the use of FMT as treatment for an 18-year-old male patient who was diagnosed with autism spectrum disorder and presented with diarrhea-predominant irritable bowel syndrome (54). The patient’s clinical symptoms were comprised of difficulty communicating with people, feelings of anxiety, impulses of destroying things, and occasional auditory hallucinations and delusions. A psychiatrist conducted the HAM-A, HAM-D, and the Symptom Checklist-90 (SCL-90), a questionnaire used to assess psychological problems, at the following post-FMT follow-ups: 1-week, 1-month, and 3-months post-treatment. Results highlighted that there was a significant improvement in the patient’s mental state at 1-week and 1-month post-treatment, however psychological symptoms reappeared at the 3-month follow-up. Interestingly, while clinical symptoms reappeared at 3-months post-treatment, the patient’s IBS symptoms maintained improvement (54).

On the contrary, Cai et al. (2019) followed the case of a 79-year-old woman diagnosed with depression and presented with loss of appetite, weight loss, constipation, introversion, and drowsiness (53). The patient received FMT via gastroscope and was followed up at 3 timepoints post-treatment: 4 days, 2- and 24-weeks. Using the Patient Health Questionnaire (PHQ-9), a questionnaire used to assess the severity of depression, patient scores improved from 21 at pre-treatment to 4 at 24-weeks post-treatment, indicating a return to average levels (53). Additionally, the patient’s weight also returned to normal and constipation symptoms had improved.

Wang et al. (2023) conducted a randomized study on the effect of combined probiotic and cerebral infarction medications, versus only cerebral infarction medication, on post-stroke depression. Patients were followed up at 6 timepoints post-treatment: 1, 7, 14, 30, 60, and 90 days. Using the HAM-D, scores were significantly lower in the probiotic intervention group, than the observation group, across all timepoints (55).

This review presented with mixed findings in terms of evaluating the longitudinal efficacy of microbial therapeutics as treatment for psychiatric disorders. Given that the majority of studies that met inclusion criteria for this review were conducted in populations with a G.I.-related illness and evaluated psychiatric well-being as a secondary outcome measure, no strong conclusions can be made regarding the longitudinal efficacy of microbial treatment use within psychiatry. That being said, the studies available do provide valuable insight into the long-term effects of microbial therapeutics and their associated relapse and/or remission rates. These results also highlight the rarity of studies conducting long-term clinical and biological follow-up assessments, which is absolutely necessary to ascertain if sustained improvements in mood – are related to lasting changes in gut microbiome composition.

Of the nine studies that assessed for depressive symptomology: the majority (n = 6) reported a lasting improvement in mood at 3-months post-treatment (47, 50, 52, 53, 55, 56); some (n = 2) reported an improvement in symptoms shortly after treatment, but the effect diminished at 3-months post-treatment (49, 54); and one study found no significant reduction in symptoms (46). Interestingly, while both Huang et al. (2022) and Lahtinen et al. (2020) found no sustained improvement in mood, they both reported a lasting reduction in GI-related symptom severity for IBS and refractory-IBS, respectively (46, 54). Huang et al. (2019) reported that the improvement in mood was more significant at 3-months than at 6-months post-treatment, with Johnsen et al. reinforcing this claim by also finding an improvement in mood at 6-months after treatment, but no sustained effect by 12-months (47, 48). On the contrary, both Xie et al. and Cai et al. documented a lasting improvement in mood at 6- and 18-months post-treatment (50, 53). Only two of the six studies that assessed for symptoms of anxiety, reported a sustained improvement in symptoms at 3-months post-treatment (47, 53), however the rest either found short-term improvement or no sustained improvement past 3 months (46, 49, 52, 54).

Based on these mixed findings and lack of consistency between results, it becomes clear that there is a critical need for more randomized, placebo-controlled, studies with a longitudinal component in order to fully assess the long-term potential of microbial therapeutics. Given the recent FDA approval of Rebyota, a fecal microbiota product delivered rectally, and Vowst, a fecal microbiota product delivered orally, there is an increasing sense of urgency for a full characterization of the long-term effects involved with gut microbiome-targeting treatments (57, 58). In efforts to accomplish this, it is pertinent to reach a consensus on the regulation and standardization across the use of microbial therapeutics. A standardization of criteria is not only warranted for creating consistency across CFU dosage ranges, frequency of application, and timing of treatment, but it’s equally as necessary for the selection of healthy donors. Without a clear definition of a ‘healthy’ microbial composition of the gut, it leaves room for discrepancies amongst studies in regard to the criteria used for selecting healthy donors. This ultimately diminishes the consistency and comparability between studies, while also posing as a possible safety risk in terms of the varying degrees of stringency across donors. Studies need to evaluate the long-term effects of these treatments, in diverse populations, with as much transparency as possible, when disclosing the reasoning behind the decisions made regarding healthy donor selection and the procedural methodology.

Given the novelty of using microbial therapeutics for psychiatric disorders, there is limited research that exists in this field, especially while considering the long-term efficacy. Though the studies included in this review contributed to a deeper understanding of the long-term effects of microbial therapeutics in the context of psychiatric well-being, some limitations still exist. One significant limitation was the rarity of longitudinal follow-ups in studies with psychiatric disorders as the main indication. All but three studies included in this review assessed non-psychiatric-focused disorders as the primary indication, and not necessarily those exclusively with psychiatric disorders. Hence, any improvement of psychiatric symptoms cannot be entirely attributed to a direct relationship between microbial composition and its influence on mood. Rather, improvements of physical symptoms may result in downstream improvements of psychiatric symptoms. This confounding limitation is evident in any bidirectional relationship, as it becomes indistinguishable if the treatment improved mood directly, or rather, improved mood through the means of improving G.I. symptoms. Many of the studies also had small sample sizes or lack of randomization and placebo-controlled design, making it difficult to generalize results and comment on the long-term efficacy rates.

Another limitation found in this review was the lack of consistency for follow-up duration across all studies. Although our search criteria maintained a minimum follow-up timepoint of 3-months in duration following treatment, it is difficult to compare and contrast multiple studies when they differ in follow-up durations. A study which utilizes a single follow-up timepoint of 18-months post-treatment, even with reported favorable results, cannot be regarded to the same standard as a study that produced favorable results at 6-months post-treatment. With the absence of consistency across data points, it is impossible to suggest that an improvement at 6-months after treatment would also yield an improvement at 18-months post-treatment.

Due to the nature of longitudinal studies in clinical settings, confounding variables and extenuating circumstances will always be a factor to consider. It is very possible that any sustained improvements in mood post-treatment, may have also been influenced by environmental stimuli and/or other lifestyle changes between visits. However, the same argument can also be made for many other psychiatric treatment options post-intervention, as environmental stimuli will always be a factor to consider, regardless of treatment choice. Given the uncertainty of whether the gut microbiome reverts back to a pre-treatment state or not, the call for multiple well-documented follow-ups, assessing microbial composition, is needed for establishing the extent and efficacy of long-term stability. In pursuance to solving this challenge, the assessment of microbial diversity at varying frequencies after initial treatment, such as 1-,3-,6-,12-, and 18-months post-FMT, would be highly beneficial. Dependent on those longitudinal findings and varying rates of response or improvement, it is then possible to gauge the need of future booster treatments. It is only through this repetitive process of long-term follow-ups and longitudinal characterization of the gut, that framework and guidelines surrounding the use of microbial therapeutics would be more concrete in its regulation.