Patient-reported treatment outcomes in ME/CFS and long COVID

Abstract

Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) and long COVID are persistent multisystem illnesses affecting many patients. With no known effective FDA-approved treatments for either condition, patient-reported outcomes of treatments may prove helpful in identifying management strategies that can improve patient care and generate new avenues for research. Here, we present the results of an ME/CFS and long COVID treatment survey with responses from 3,925 patients. We assess the experiences of these patients with more than 150 treatments in conjunction with their demographics, symptoms, and comorbidities. Treatments with the greatest perceived benefits are identified. Patients with each condition who participated in the study shared similar symptom profiles, including all the core symptoms of ME/CFS, e.g., 89.7% of ME/CFS and 79.4% of long COVID reported postexertional malaise (PEM). Furthermore, treatment responses between these two patient groups were significantly correlated (R² = 0.68). Patient subgroups, characterized by distinct symptom profiles and comorbidities, exhibited increased responses to specific treatments, e.g., a POTS-dominant cluster benefiting from autonomic modulators and a cognitive-dysfunction cluster from CNS stimulants. This study underscores the symptomatic and therapeutic similarities between ME/CFS and long COVID and highlights the commonalities and nuanced complexities of infection-associated chronic diseases and related conditions. While this study does not provide recommendations for specific therapies, in the absence of approved treatments, insights from patient-reported experiences provide urgently needed real-world evidence for developing targeted patient care therapies and future clinical trials.

Keywords: ME/CFS; long COVID; patient-reported outcomes; real-world evidence; treatments.

Fig. 1.

Most troubling symptoms in ME/CFS and long COVID patients. The bar graph shows the frequency of symptoms reported by patients with ME/CFS (blue) and long COVID (red), ordered based on the frequency in long COVID patients.

Fig. 2.

Patient-reported outcomes of treatments in ME/CFS and long COVID. (A) The leading treatments and categories that contributed to improving patients’ overall conditions as reported in self-assessments. The data within parentheses next to each treatment indicate the number of responses provided on the treatment, the false discovery rate adjusted P-value compared to the reference treatment, and the Net Assessment Score (NAS). The stacked bar plots show the percentages of patient responses related to the perceived treatment effect of each treatment, spanning from “much worse” to “much better”. Orange shading denotes the percentage of patients who reported a positive treatment effect—with darker orange indicating a stronger positive effect—whereas blue shading indicates adverse side effects, with darker blue representing more significant negative effects. Vitamin C (oral, nonliposomal) was used as the reference treatment. Of all the treatments investigated, Graded exercise therapy (GET) received the lowest NAS. (B) Impact of treatments on patient symptoms as reported in self-assessments. For each of the core symptoms, shown in a gradient scale is the proportion of respondents of a treatment who also reported the impact of the treatment improving the specific symptom.

Fig. 3.

Comparison of treatment effectiveness reported in ME/CFS and long COVID. Shown is the scatter plot of the NAS of treatments in ME/CFS (x-axis) compared to long COVID (y-axis), with each circle symbolizing a treatment. Red circles represent treatments that showed significance when compared to the reference (Vitamin C; oral, NOT liposomal), and gray circles indicate otherwise. The size of a circle represents the total number of respondents to the treatment.

Fig. 4.

Patient subgroups with distinct symptom profiles and treatment efficacies. (A) The heatmap displays the percentages of reported symptoms and comorbidities within each patient cluster. Cluster 1 had the highest prevalence of symptoms and comorbidities, Cluster 2 was predominated by postural orthostatic tachycardia syndrome (POTS) issues, Cluster 3 was characterized by cognitive symptoms like brain fog and sleep disturbances, and an increased reporting of pain, and Cluster 4 had the mildest symptoms. (B) The heatmap shows the comparative effectiveness of treatments across the four distinct patient clusters. The treatments listed are consistent with those reported in Fig. 2, and the NAS for each treatment in patients of each cluster is shown. The colors indicate treatment efficacies within each cluster, with warmer colors (reds and oranges) representing higher positive impact, and cooler colors (blues) representing lower positive impact. Gray denotes insufficient feedback (<20) for reliable NAS calculation.