Study design
This study was designed as a multicenter, randomized, double-blind, placebo-controlled clinical trial that lasted for 14 days, including 2 days of screening period, 14 days of treatment period, and the 14th day follow-up period. Screening was undertaken within two days before enrollment to assess eligibility and collect baseline data regarding basic medical history, combined medications, laboratory and auxiliary examinations [novel coronavirus immunoglobulin G (IgG) antibody and inflammatory factors], signs and symptoms, and assessment scales. Eligible participants were randomized into the SLKF granules group and the control group at a ratio of 1:1. Participants' medication, signs and symptoms, rating scales, and laboratory biochemical examinations on the scheduled day for the follow-up period were recorded. A workflow diagram and schedule are presented in Fig. 1 and Table 1, respectively. The trial protocol was designed following the Recommendations for Interventional Trials Traditional Chinese Medicine (SPIRIT-TCM) Extension 2018 Statement checklist. (Online Supplemental Appendix 1).
Participants
Participants, diagnosed as outpatients or inpatients, were recruited and screened in the following ten clinical research sub-centers: Hunan Provincial Hospital of Integrated Traditional Chinese and Western Medicine (the Affiliated Hospital of Hunan Academy of Chinese Medicine), the Second Xiangya Hospital of Central South University, the First Hospital of Hunan University of Chinese Medicine, the Second Hospital of Hunan University of Chinese Medicine, Hunan Province Directly Affiliated TCM Hospital, the First Affiliated Hospital of Shaoyang University, the First Traditional Chinese Medicine of Changde, Hengyang Hospital of Traditional Chinese Medicine, and Zhangjiajie Hospital of Traditional Chinese Medicine. This trial started with recruitment from January 2024 to December 2024.
Most participants were recruited through advertisements and posters in ten selected clinical research subcenters. Other participants diagnosed with convalescent COVID-19 were invited by telephone through the research team or clinical observers. All participants were required to be negative for novel coronavirus antibody or nucleic acid detection; IgG antibody levels were elevated [24] and re-screened by a specialist. Each participant signed an informed consent form (ICF) indicating that they fully understood the clinical study information concerning the trial.
Diagnostic Criteria
Diagnostic Criteria for the Convalescent COVID-19
According to the 'Diagnosis and Treatment of COVID-19 (trial tenth edition)' and the Expert Guidelines on Health Management for Convalescent COVID-19 Patients (first edition)'[14–15], Convalescent COVID-19 refers to those who have been infected, met any of the following criteria, and had significantly improved other symptoms.
(1) Two consecutive nucleic acid detections were negative, and the cycle threshold (Ct) values were ≥ 35.
(2) Negative antigen test results on three consecutive days.
(3) The level of novel coronavirus-specific IgG antibodies in convalescence was four times or higher than that in the acute stage.
(4) In the absence of antipyretics, fever symptoms subsided for more than 24 h.
(5) Pulmonary imaging demonstrated that acute exudative lesions were significantly improved and could be converted to oral drug therapy without complications necessitating further treatment.
Diagnostic criteria for TCM syndrome
The TCM symptom scoring table for lung-spleen qi deficiency syndrome was formulated following 'the Guiding Principles of Clinical Research on the Treatment of Diarrhea with New Chinese Medicine (2002 edition)'[25], 'the Diagnostic criteria for TCM syndromes of coronavirus disease 2019 (trial edition)' [26], and 'the Diagnostic criteria of TCM Syndromes of Chronic Obstructive Pulmonary Disease (2011 edition)' [27]. The cardinal and minor symptoms included in the scale were as follows:
Cardinal symptoms of cough without or less phlegm, fatigue, panting, shortness of breath, torpid intake, and sloppy stool, and minor symptoms of spontaneous sweating and aversion to wind, low-grade fever, abdominal distension, cognitive impairment, insomnia, pale white complexion, dizziness, and tinnitus. Five or more points were allocated for the cardinal and minor symptoms, combined with fat tongue, teeth-marked tongue, pale tongue, white or white greasy tongue coating, sunken, fine, moderate, and weak pulse. If the above conditions were met, it was identified as lung-spleen qi deficiency syndrome.
Inclusion criteria
(1) Male and female patients aged 18–75 years.
(2) Participants had a confirmed or physician-suspected SARS-CoV-2 infection at least six months earlier.
(3) Diagnosed with convalescent COVID-19.
(4) Diagnosed with lung-spleen qi deficiency syndrome.
(5) The participants had no short-term migration intentions and agreed to cooperate with the follow-up.
(6) Provision of written informed consent.
Exclusion criteria
(1) Severe liver dysfunction: ALT > 2-time normal value upper limit or AST > 2-time the normal upper limit or severe kidney dysfunction Scr > 1.5-time normal value the upper limit).
(2) Severe pulmonary dysfunction [(forced expiratory volume in one second (FEV1%) of predicted value, including obstruction or restriction: severe: 35–49%, extremely severe: A value of less than 35% for FEV1% or a value of less than 40% for carbon monoxide diffusing capacity. For all measurements using z-values, severe impairment was < − 4.1] [28].
(3) Underlying blood disorders, acute rheumatic connective tissue diseases, or acute pulmonary conditions.
(4) Pregnancy, planning to become pregnant, or breastfeeding.
(5) History of drug rashes or anaphylactic reactions to medications.
(6) Having participated in another clinical study within three months.
(7) Combined with severe mental illness.
(8) Recently used traditional Chinese medicine or proprietary Chinese medicine with an identical effect.
Withdrawal criteria
(1) The patient's condition worsened or relapsed during the administration period, and other treatment measures are required.
(2) Subjects did not meet the inclusion criteria and were mistakenly included in the trial or met the trial's exclusion criteria.
(3) Subjects with poor compliance were not recommended to withdraw from the trial but were no longer receiving drug treatment and were lost to follow-up.
(4) Voluntary withdrawal.
(5) Subjects experienced other complications, and it was inappropriate to continue the trial during the clinical trial [29–30].
Randomization and allocation
This trial adopted a stratified block randomization method. First, an appropriate block length was selected (block size = 4), and random assignment was conducted using SAS (version 9.4) statistical software developed by statisticians from the Drug Clinical Evaluation Research Center of Central South University. A total of 154 random sequences were generated for each group; specifically, a randomized coding table (No. 001-154) was developed. The subjects in each group were randomly divided into an intervention group and a control or placebo group at an allocation ratio of 1:1. Randomized codes were used as drug numbers, which were kept in an opaque sealed envelope and taken to Hinye Pharmaceutical Co., Ltd. (Changsha Hunan, China) for unified drug label production. The identification code for each subject was similar to that of the number of drugs.
Blinding
Doctors, patients, outcome investigators, and statisticians were blinded to the patient's treatment assignments. The statisticians completed the randomized codes, and the staff of Hinye Pharmaceutical Co., Ltd. affixed the drug number label to the eye-catching position of the external drug packaging according to the drug number label corresponding to the formed random code. The record of the blinding process, the blind record, was kept as a clinical trial document. The contents of the blinding record included the preparation of the sponsor's drugs, drug packaging, usage, storage and assignment, generation of randomized codes, the subject's drug box production, emergency letter, test report of SLKF granules and placebo, storage of blinding codes, rules for breaking the blind law, and dispensing numbers of each research center. Blinded coding, the initial randomization number, and block length were sealed in two copies. For proper storage, the two copies were handed to the clinical trial group leader, the Department of Drug Clinical Trial Administration at Hunan University of Chinese Medicine, and the sponsor. The blinding codes were kept strictly confidential and remained sealed until statistical analysis at the end of the trial, only if serious adverse events (SAEs) or emergencies occurred.
Intervention
Treatments
Multicenter subjects satisfying all criteria were assigned (1:1) randomly into an intervention group and a control group. All participants received health management guidance (maintaining good personal and environmental hygiene, balanced nutrition, moderate exercise, adequate rest, and avoiding fatigue). Subjects with chronic underlying diseases were administered corresponding treatments. The intervention group was treated with SLKF granules orally (once per bag, 16.9 g, twice daily). Subjects in the control group were treated with a placebo at the same dosage, which had an identical appearance and nearly similar taste to SLKF granules. If necessary, routine treatment, including cough relief and other symptomatic support, was administered. The treatment for the two groups lasted 14 days. Each center was assigned a clinical coordinator (CRC) responsible for allocating drugs based on the recruitment sequence before filling out the dispensing record form.
Preparation of drugs
SLKF granules were composed of traditional TCM formula granules that met the quality standards according to the proportion of prescriptions. Each package was equivalent to decoction: Dangshen (Codonopsis Radix) 3.75 g (9.8%), Huangqi (Astragali Radix) 3.75 g (9.8%), Fuling (Poria) 3.75 g (9.8%), Baizhu (Atractylodis Macrocephalae Rhizoma) 2.25 g (5.88%), Maidong (Ophiopogonis Radix) 3.75 g (9.8%), Guizhi (Cinnamomi Ramulus) 2.25 g (5.88%), Tianma (Gastrodiae Rhizoma) 3 g (7.84%), Jiegeng (Platycodonis Radix) 2.25 g (5.88%), Zhiqiao (Aurantii Fructus) 2.25 g (5.88%), Gegen (Puerariae Lobatae Radix) 2.25 g (5.88%), Baihe (Lilii Bulbus) 3 g (7.84%), Baishao (Paeoniae Radix Alba) 2.25 g (5.88%), Chuanxiong (Chuanxiong Rhizoma) 2.25 g (5.88%) and Zhigancao (Glycyrrhizae Radix Et Rhizoma Praeparata Cum Melle) 1.5 g (3.92%). An SBH-200 three-dimensional mixer, BW-1500 mixer, or EYH-2000 two-dimensional motion mixer was used for the operation. Following the prescription, the formula particles of each Chinese medicine were placed in the total mixer, mixed for 10 min, and discharged. A DS-100 automatic vertical high-speed trilateral sealing particle packaging machine was used for the subpackaging. The qualified composite film bag was packaged and sealed according to the packaging specifications (16.9 g/package). Finally, the label and box were used to obtain the finished drug.
The placebo was developed based on caramel color liquid, maltodextrin, and purified water at a weight ratio of 1:20:25. The caramel color liquid and maltodextrin were dissolved in purified water in turn, mixed into thick paste, and dried to dry paste at 90℃. Subsequently, the dry paste was sieved to remove fine particles through an 60 mesh sieve, and discharged a pale yellow powder. SBH-200 three-dimensional mixer, BW-1500 mixer, or EYH-2000 two-dimensional motion mixer were used for total mixing. After mixing, 1/3 of the powder was first mixed with Chinese medicine powder flavor, bitter agent and sucralose for 10 min, then 1/3 of the powder was added for 10 min, and finally the remaining 1/3 of the powder was added for 5 min. The GFS-200/ LGP200 dry granulator was used to granulate, the powder was pressed into a strip, and the particles were made after passing through a 12–14 stainless steel mesh sieve, whose package, color, and flavor were consistent with those of the SLKF granules.
Thin-layer chromatography (TLC) and high-performance liquid chromatography (HPLC) were used to characterize the primary active ingredients of SLKF granules. Compounds, such as gastrodin (C13H18O7), p-hydroxybenzyl alcohol (C7H8O2), parishin E (C19H24O13), parishin B (C32H40O19), parishin C (C32H40O19) and parishin A (C45H56O25), were the primary bioactive components of Tianma in SLKF granules with a concentration of 43.0 ~ 80.0 mg/g. 2-acetate regaloside A (C20H26O11), was identified as the primary active ingredient of Baihe with a concentration of 3.0–8.5 mg/g. Ferulic acid (C10H10O4), the primary active ingredient of Chuanxiong was 1.5–4.5 mg/g. Platycodin D (C57H92O28), the primary active ingredient of Jiegeng was 1.2–4.0 mg/g. Paeoniflorin (C23H28O11), the primary active ingredient of Baishao was 65.0–137.0mg/g. Puerarin (C21H20O9), the primary active ingredient of Gegen was 55.0–110.0 mg/g. Neochlorogenic acid (C16H18O6), chlorogenic acid (C16H18O6) and cryptochlorogenic acid (C16H18O9 ), the primary active components of Baizhu were 0.12–0.95mg/g. It also contains fructose (C6H12O6) and sucrose (C12H22O11), and the total amount is 30.0–155.0mg/g. Each gram of Zhiqiao granules contained neohesperidin (C28H34O15) 49.0–101.0mg and naringin (C27H32O14) 66.0–148.0mg. Cinnamic acid (C9H8O2), the primary active ingredient of Guizhi was 4.5–18.5mg/g. Calycosin-7-glucoside (C22H22O10), the primary active ingredient of Huangqi was 1.20–3.50mg/g. Lobetyolin (C20H28O8), the primary active ingredient of Dangshen was 0.08–0.03mg/g. Ophiopogon methylflavanone A (C19H18O6) and Ophiopogon methylflavanone B (C19H20O5), the primary active ingredients of Maidong were 0.010–0.100 mg/g. Each gram of licorice granules contained Liquiritin (C21H22O9) 6.5–23.0mg and glycyrrhizic acid (C42H62O16) 12.9– 60.0mg.
Prohibited drugs
Nonsteroidal anti-inflammatory drugs, steroids, and antibiotics were prohibited. The Chinese medical treatments used to treat those residual symptoms were prohibited during the study period.
Outcome measures
Determination of core outcome indicators for the effectiveness trial based on Core Outcome Measures in Effectiveness Trials (COMET) (https://www.comet-initiative.org/)[31–33].
Primary outcome
Improvement of lung-spleen qi deficiency syndrome using the therapeutic effective rate on day 14 and total clinical symptom score on days 0 and 14.
Secondary outcomes
(1) Improvement in the fatigue self-assessment scale (FAS) on days 0 and 14.
(2) Changes in the pain visual analog scale (VAS) scores on days 0 and 14.
(3) Improvement of the Pittsburgh sleep quality index (PSQI) on days 0 and 14.
(4) Improvement of mini-mental state examination (MMSE) on days 0 and 14.
(5) Improvement of hospital anxiety and depression scale (HADS) on days 0 and 14.
(6) Changes in TCM syndrome scores on days 0 and 14.
(7) Changes in C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), interleukin-6 (IL-6) on days 0 and 14.
Safety outcomes
(1) Adverse events (AEs) related to study treatment.
(2) Blood routine tests [red blood cell (RBC), white blood cell (WBC), hemoglobin (HGB), and platelet count (PLT)], urine routine tests (WBC, RBC, and urine glucose), and stool routine tests (qualitative + occult blood).
(3) Liver function [alanine aminotransferase (ALT), aspartate transaminase (AST), gamma-glutamyl transferase (GGT), total bilirubin (TBiL), and alkaline phosphatase (ALP), blood urea nitrogen (BUN), serum creatinine (Scr), urine acid (UA), estimated glomerular filtration rate (eGFR), and electrocardiogram (ECG).
These assessments were conducted before enrollment and on the 14th day after enrollment.
SLKF granules, Shenlingkangfu granules; FAS, fatigue self-assessment scale; VAS, pain visual analog scale; PSQI, Pittsburgh sleep quality index; MMSE, mini-mental state examination; HADS, hospital anxiety and depression scale; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; IL-6, interleukin-6; AEs, adverse events; ECG, electrocardiogram.
Table 1
| Items | Screening Period | Treatment Period | Follow-Up Period |
| -2nd Day ± 2 Days | 1st-14th Day | 14th Day |
| ENROLMENT: |
| Informed consent | X | | | |
| Eligibility screen | X | X | | |
| Demographic information | X | | | |
| Basic medical history | X | | | |
| Combined medication | X | X | | X |
| Random assignment | | X | | |
| Novel coronavirus IgG antibody detection | X | | | |
| INTERVENTIONS: |
| SLKF granules | | | X | |
| Placebo | | | X | |
| ASSESSMENTS: |
| Therapeutic effective rate | | | | X |
| Total clinical symptom score | | X | | X |
| FAS | | X | | X |
| VAS | | X | | X |
| PSQI | | X | | X |
| MMSE | | X | | X |
| HADS | | X | | X |
| TCM syndrome score | | X | | X |
| CRP, ESR, IL-6 | | X | | X |
| Three routine tests | | X | | X |
| Liver and kidney function | | X | | X |
| ECG | | X | | X |
| AEs | | X | | X |
SLKF granules, Shenlingkangfu granules; FAS, fatigue self-assessment scale; VAS, pain visual analogue scale; PSQI, Pittsburgh sleep quality index; MMSE, mini-mental state examination; HADS, hospital anxiety and depression scale; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; IL-6, interleukin-6; three routine tests, routine blood tests, urine routine tests, and stool routine tests; ECG, electrocardiogram; AEs, adverse events.
Assessments
(1) The effectiveness of clinical treatment
Following the 'Guidelines for Clinical Research of New Chinese Medicinal Drugs (2002 edition)' [25], the efficacy of TCM syndromes was determined using the following nimodipine method calculation formula: [(pre-treatment score ˗ post-treatment score) / pre-treatment score] × 100%.
a) Clinical cure: The disappearance or near-complete disappearance of TCM clinical symptoms and signs, with a reduction in syndrome score of ≥ 95%.
b) Significant effect: Significant improvement in TCM clinical signs and symptoms, with a reduction in syndrome score of ≥ 70%.
c) Effective: Improvement in TCM clinical symptoms and signs, with a reduction in syndrome score of ≥ 30%.
d) Ineffective: No significant improvement in TCM clinical symptoms and signs, or even worsening, with a reduction in syndrome score of < 30%.
The total score for clinical symptoms included the following 16 items: fatigue/tiredness, pain/discomfort (muscle, joint, headache, and chest pain), cough or sputum production, sleep disturbances, excessive sweating, decreased physical stamina, taste disorders, smell disorders, and others. Based on the patient's actual condition and experience, they indicated the presence or absence of these symptoms. 'Absent' was recorded as 0 points and 'present' as 1 point, with the total amount representing the overall clinical symptom score. A higher score indicated a more severe manifestation of 'long COVID' symptoms during the recovery phase.
(2) FAS
The FAS, VAS, PSQI, MMSE, HADS, and TCM syndrome scores were evaluated at baseline and 14 days after enrollment.
The FAS employed a Likert-type scale scoring method consisting of 10 items to assess chronic fatigue symptoms. Each item was rated on a 5-point scale ranging from 1 ('never') to 5 ('always') [34]. The total score ranged from 10 to 50, with a score of 10 representing the lowest level of fatigue and 50 indicating the highest level. Items 4 and 10 were reverse-scored in comparison with the other eight items.
(3) VAS
The VAS utilized a 10 cm line segment marked with 10 increments, with one end labeled '0 points' and the other '10 points.' Patients were instructed to mark a point along the line that corresponded to the intensity of their pain at that moment (using a dot or an '✕') [35]. A measurement of 0 cm (0 points) indicated no pain; 1–3 cm (1–3 points) represented mild pain, did not affect work or daily life; 4–6 cm (4–6 points) signified moderate pain, impacting work but not daily activities; 7–10 cm (7–10 points) indicated severe pain, substantially affecting both work and life.
(4) PSQI
The PSQI is used to evaluate a patient's sleep quality over the past month. It comprises seven components: sleep quality score, time to fall asleep score, sleep duration score, sleep efficiency score, sleep disturbances score, use of sleeping medication score, and daytime dysfunction score. Each component was scored from 0 to 3, with the total PSQI score ranging from 0 to 21. The scoring thresholds were as follows: 0–5 points (very good sleep quality), 6–10 points (fair sleep quality), 11–15 points (average sleep quality), and 16–21 points (very poor sleep quality) [36]. The PSQI score is inversely related to sleep quality, with a higher score indicating poorer sleep quality.
(5) MMSE
The MMSE consists of 12 items that assess a participant's intellectual status and degree of cognitive impairment across dimensions, including orientation, memory, attention and calculation, recall, and naming ability [37]. The highest possible score is 30, with scores between 27 and 30 indicating normal cognitive functioning and scores below 27 suggesting cognitive impairment.
(6) HADS
The HADS consists of 14 items, with seven items assessing depression and seven items evaluating anxiety. There are six reverse-scored items, five of which are included in the depression subscale and one in the anxiety subscale. Each item is rated on a 4-level scale from 0 to 3, with scores of 0 to 3 assigned based on the absence, mild, moderate, and severe impact on the subject, respectively[38]. A total score ranging from 0 to 7 indicated no symptoms, a score from 8 to 10 suggested the possible presence of anxiety or depression symptoms, and a score from 11 to 21 confirmed the presence of anxiety or depression symptoms.
(7) TCM syndrome score
The collection of TCM symptom scores included the following main symptoms: cough with little or no phlegm, fatigue and weakness, wheezing and shortness of breath, decreased appetite, and loose stools. These symptoms were rated based on severity, with scores of 0, 2, 4, and 6 for none, mild, moderate, and severe, respectively. Additionally, secondary symptoms, including spontaneous sweating and aversion to wind, low-grade fever, abdominal distension, cognitive impairment, insomnia, pallor, and dizziness with tinnitus, were evaluated and assigned scores of 0, 1, 2, and 3 according to their severity [25].
(8) Blood test
Venous blood samples were collected from all participants to determine C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), and interleukin-6 (IL-6) levels.
Patient visit and data collection
Baseline data
Baseline data included general demographic data, past medical history, course of disease, treatment history, drinking history, participation in clinical trials, total clinical symptom score, fatigue, sleep, pain, cognition, depression scales, TCM syndrome score, and safety evaluations. The main components included the vital signs of the subjects (body temperature, blood pressure, respiration, and heart rate), main signs and symptoms, personal history and medication, total score of clinical symptoms, FAS, VAS, PSQI, MMSE, HADS, TCM syndrome score, CRP, ESR, IL6, routine blood, urine and stool tests, liver and kidney function, and ECG.
Assessment data
On the 14th day of treatment, the symptoms and signs, medication, FAS, VAS, PSQI, MMSE, HADS, a total score of clinical symptoms, TCM syndrome score, CRP, ESR, and IL6 were assessed. Laboratory examination results were collected, including routine blood, urine, and stool tests, liver and kidney function, and ECG. The onset time, severity, duration, effective measures, outcomes of AEs, completion of trials, and time of discontinuation of trials were recorded in the case report form (CRF), including an analysis of whether AEs were associated with clinical drug trials.
Indicators detection
During the screening and treatment periods, all subjects were required to take 3–5 mL of peripheral venous blood in the morning on an empty stomach and then left to stand for 1–2 h. Whole blood, serum, and plasma supernatants were collected (extracted after centrifugation at 3,000 rpm for 10–15 min) for routine blood tests, liver and kidney function, CRP, ESR, and IL6. Three tubes of urine from each subject's midstream specimen were collected for urine sediment microscopy and routine urine analysis. An appropriate amount of stool samples were kept for routine stool and occult blood tests. Laboratory biochemical examinations were performed using a cobas8000 (702/502), cobasU601/701 Roche automatic biochemical analyzer, and SYSMEX-Xi 9000 blood cell analyzer.
Adverse events
The protocols for handling specific AEs and SAEs are described in Supplemental Appendix 2. Participants were asked if they had experienced any AEs at each study visit. AEs were assessed by a study physician and reported to the Data Safety and Monitoring Board. SAEs, including death, life-threatening conditions, inpatient hospitalization or prolongation of existing hospitalization, persistent or significant disability or incapacity, congenital anomaly, or birth defect [39], were reported to the Ethics Review Committee of Hunan Academy of Chinese Medicine Affiliated Hospital within 24 h after acknowledgment. For mild AEs, the patient was reassured to continue taking medication following the protocol [40–41].
Sample Size
The determination of sample size in a superior clinical trial design with qualitative variables is calculated using the following formula:
$$\:\text{n}=\frac{{{\pi\:}}_{\text{t}}\times\:\left(1-{{\pi\:}}_{\text{t}}\right)+{{\pi\:}}_{\text{c}}\times\:\left(1-{{\pi\:}}_{\text{c}}\right)}{{\left({{\pi\:}}_{\text{t}}-{{\pi\:}}_{\text{c}}-\varDelta\:\right)}^{2}}\times\:{\left({{\mu\:}}_{{\alpha\:}/2}+{{\mu\:}}_{{\beta\:}}\right)}^{2}$$
A clinical trial investigating the effects of TCM in treating patients with convalescent COVID-19 and lung and spleen qi deficiency syndrome demonstrated that the medicine improves patients' symptoms, including fatigue, shortness of breath, cough, and poor appetite. After two weeks of treatment, the improvement rate for individual symptoms in the TCM group ranged from 85.4–89.7%, whereas that in the placebo group ranged from 42.2–64.7% [40]. Assuming a clinical effective rate of 90% for the experimental group (treated with SLKF Granules) and 65% for the control group (receiving a placebo simulation of SLKF Granules), with a significance level of 5%, an anticipated dropout rate of 15%, setting α = 0.025 (one-sided) and β = 0.20 (one-sided), and Δ = 5%, the sample size calculation indicated that at least 63 participants were required for each group. Considering the 15% dropout rate, 154 participants were needed for both groups.
Statistical Methods
All statistical analyses were performed using SAS software (version 9.4). For continuous variables, the sample size, mean, standard deviation, median, and minimum and maximum values were calculated. Intergroup comparisons for the treatment groups were performed using one-way analysis of variance or rank-sum tests. For counting data, the frequency and composition ratio were calculated, with intergroup comparisons for the treatment groups using the chi-squared or Fisher's exact test. Regarding medication adherence and analysis of concomitant medication, the percentage of participants with adherence rates between 80% and 120% was calculated, and group differences were compared using the chi-square test or Fisher's exact probability method. The percentage of participants taking concomitant medications was also calculated, and differences between groups were compared using the chi-square test or Fisher's exact probability method. Additionally, an analysis based on the WHO anatomical therapeutic chemical (ATC) classification for concomitant medications was conducted, employing either the chi-square test or Fisher's exact probability method to compare group differences. The comparison and analysis of the effectiveness indicators employed non-parametric tests or an analysis of covariance. If baseline differences existed between the two groups, baseline measurements were adjusted as covariates before conducting intergroup comparisons. The safety analysis primarily involved assessing the AE incidence between the groups, abnormal laboratory data, electrocardiograms, and physical examination results before and after treatment.
The full analysis set was conducted for the therapeutic indices of all subjects included in the study. Statistical analysis in accordance with the per protocol set (PPS) was conducted for the therapeutic indices of subjects completing the drug intervention on the 14th day. The full analysis set (FAS) used the last-observed-carried forward method to adjust for missing data and complete the intention-to-treat analysis. Safety analysis was performed using the safety analysis set. All statistical tests were two-sided. P ≤ 0.1 was considered a statistical difference, while P ≤ 0.05 was considered a statistically significant difference.
Quality Control and Data Monitoring
Quality assurance
First, all participating researchers and laboratories were required to adhere to the laboratory's quality control measures and drug batch quality control following standard operating procedures and quality control protocols. Each participating unit provided the 'normal value range for laboratory tests' specific to their facility, and any changes during the trial were promptly documented.
Second, to control for participant compliance, a combination of drug counting and questioning methods was used to monitor adherence to the study medication. Adequate explanations were provided to participants, and follow-ups were strengthened to ensure good compliance among the subjects. The participants received training before the trial to ensure that they fully understood the informed consent. When distributing diary cards, researchers carefully explained the filling requirements, methods, and return procedures, as well as educated participants on potential adverse reactions to the trial medication and actions to take if such reactions occur.
Finally, in terms of investigator control, training was held before the clinical trial was launched to ensure that the investigators fully understood the clinical trial protocol and the specific connotations of each indicator. Moreover, the consistency of the quantitative standard of symptoms and signs was tested, the consistency of the TCM syndrome was trained and evaluated, and the investigators' subjective symptom description was aimed to avoid inducing or hinting. The objective indicators were checked according to specified time points and methods. Attention was paid to observing adverse reactions or unexpected toxic side effects, followed by follow-up assessments. Research assistants filled out the CRF following the researcher's manual in each center, requiring them to record the contents of the form accurately and thoroughly to ensure the authenticity and reliability of the CRF content. All observations and findings in the clinical trial were verified to ensure the reliability of the data, and all conclusions in the clinical trial were based on the original data.
Data management
In terms of data monitoring and management, this trial employed CRFs for data collection and utilized Epidata for data management. An independent third party, separate from the CRC, was entrusted with data verification for the CRFs. The database was provided by the Clinical Evaluation Research Center of Central South University, and the data were entered into the database by two personnel to ensure accuracy. Subsequently, statistical personnel from the center crosschecked the raw data. Any identified issues were promptly questioned and returned for modification and enhancement. Once the verification was completed and accurate, data management and statistical analysis were performed.
Confidentiality
All study-related information was stored on secure network drives or servers. Participants were allocated study identification (ID) numbers, and a master file linking the study ID and personal information was saved separately to protect their privacy.
Access to data
The study principal investigator, co-investigators, CRC, research assistants, and statistician have access to the collected data.
Ethics approval and study registration
This study was approved by the Ethics Review Committee of Hunan Academy of Chinese Medicine Affiliated Hospital (reference number: 2023 − 144), any protocol modifications will be reported to the Ethics Review Committee.
The research staff obtained consent from interested participants through ten hospitals in Hunan province in China, an outpatient medical record platform, an inpatient case system, and recruitment advertisements. Participants received information regarding the ICF and had the opportunity to discuss the trial specifics and meet with the CRC or research assistant before deciding to participate. The ICF is provided in the Online Supplemental Appendix 3.
The trial was subjected to inspections or audits by the Ethics Review Committee of Hunan Academy of Chinese Medicine Affiliated Hospital and Hunan Provincial Health Commission to determine whether research activities were conducted following the protocol, Good Clinical Practice, and guidelines of the International Conference on Harmonization. Under the identifier ChiCTR2400080348, the research protocol has been filed in the Clinical trials registry.