Homeopathy for Covid-19 in Primary Care: A structured summary of a study protocol for a randomized controlled trial

Objectives To investigate the effectiveness and safety of homeopathic medicine Natrum muriaticum (LM2) for mild cases of COVID-19 in Primary Health Care. Trial design A randomized, two-armed (1:1), parallel, placebo-controlled, double-blind, clinical trial is being performed to test the following hypotheses: H0: homeopathic medicines = placebo (null hypothesis) vs. H1: homeopathic medicines ≠ placebo (alternative hypothesis) for mild cases of COVID-19 in Primary Care. Participants Setting: Primary Care of São Carlos – São Paulo – Brazil. One hundred participants aged 18 years or older, with Influenza-like symptoms and a positive RT-PCR for SARS-CoV-2. Willingness to give informed consent and to comply with the study procedures is also required. Exclusion criterium: severe acute respiratory syndrome. Intervention and comparator Homeopathy: 1 globule of Natrum muriaticum LM2 diluted in 20 mL of alcohol 30% and dispensed in a 30 ml bottle. Placebo: 20 mL of alcohol 30% dispensed in a 30 ml bottle. Posology: one drop taken orally every 4 hours (6 doses/day) while there is fever, cough, tiredness, or pain (headache, sore throat, muscle aches, chest pain, etc.) followed by one drop every 6 hours (4 doses/day) until the fourteenth day of use. The bottle of study medication should be submitted to 10 vigorous shakes (succussions) before each dose. Posology may be changed by telemedicine, with no break in blinding. Study medication should be maintained during home isolation. According to the Primary Care protocol, the home isolation period lasts until the 10th day after the appearance of the first symptom, or up to 72 hours without symptoms. Main outcomes The primary endpoint will be time to recovery, defined as the number of days elapsed before all COVID-19 Influenza-like symptoms are recorded as mild or absent during home isolation period. Secondary measures are recovery time for each COVID-19 symptom; score of the scale created for the study (COVID-Simile Scale); medicines used during follow-up; number of days of follow-up; number of visits to emergency services; number of hospitalizations; other symptoms and Adverse Events during home isolation period. Randomisation The study Statistician generated a block randomization list, using a 1:1 ratio of the two groups (denoted as A and B) and a web-based tool (http://www.random.org/lists). Blinding (masking) The clinical investigators, the statistician, the Primary Care teams, the study collaborators, and the participants will remain blinded from the identity of the two treatment groups until the end of the study. Numbers to be randomised (sample size) One hundred participants are planned to be randomized (1:1) to placebo (50) or homeopathy (50). Trial Status Protocol version/date May 21, 2020. Recruitment is ongoing. First participant was recruited/included on June 29,2020. Due to recruitment adaptations to Primary Care changes, the authors anticipate the trial will finish recruiting on April 10, 2021. Trial registration COVID-Simile Study was registered at the University Hospital Medical Information Network (UMIN - https://www.umin.ac.jp/ctr/index.htm) on June 1st, 2020, and the trial start date was June 15, 2020. Unique ID: UMIN000040602. Full protocol The full protocol is attached as an additional file, accessible from the Trials website (Additional file 1). In the interest in expediting dissemination of this material, the familiar formatting has been eliminated; this Letter serves as a summary of the key elements of the full protocol. Supplementary Information The online version contains supplementary material available at 10.1186/s13063-021-05071-5.


INTRODUCTION
Nine months after the first COVID-19 reported case on February 26 th , Brazil is rapidly approaching 7.5 million cases and 190 thousand deaths 1   he considered to be his "most perfected method" 8 : repeated (as needed) doses of the "best chosen" homeopathic medicine in liquid form (so that it could be succussed before each dose 9 ), prepared trough the fifty-millesimal dynamization 10 and used in ascending potencies (when needed), beginning with the lowest degrees 6 .
Fifty-millesimal or Quinquagintamillesimal (LM or Q) potencies are prepared by trituration of the raw material (in three 1:100 steps), followed by consecutive 1:50,000 succussed dilutions. Therefore, a LM1 potency corresponds to 5x10 -10 fraction of the raw material (LM2 = 2.5 x 10 -15 , LM3 = 1.25 x 10 -20 , LM4 = 6.25 x 10 -24 etc.). Hahnemann's 18th century dynamizing or potentizing, i.e., the mechanical action upon particles of the raw material, separated from each other through the addition of an indifferent medium, dry (for trituration) of fluid (for dilution) 11 , have been recognized as a manual method to produce nanoparticles of the source material 12 .
According to Hahnemann's instructions, the finding of a homeopathically suitable remedy for an epidemic disease requires the examination of several patients to identify the peculiarities of the collective disease, i.e., the "more marked and special symptoms which are peculiar to but few diseases and of rarer occurrence, at least in the same combination" 13 . General symptoms (loss of appetite, headache, debility, sleeplessness, discomfort, etc.), on the other hand, demand but little attention, if they cannot be more accurately described and present no peculiarities, "as symptoms of such a general nature are observed in almost every disease and from almost every drug" 14 .
When COVID-19 reached São Carlos, we had no patients to examine. A literature search at that time led us to the interesting findings of the German virologist Hendrik Streeck: ca. of 2/3 of more than 100 patients affected by COVID-19 had described a long-lasting (many days) loss of smell and taste 15 .
Coincidentally, Hahnemann himself came across an epidemic fever in 1809 and ageusia was one of the characteristic symptoms stressed by him: "on taking any food, the sense of taste appears as if extinguished". Ageusia is also a pathogenetic symptom of the two medicines Hahnemann considered most appropriate to that epidemic fever: Nux vomica and Arsenicum album, this one chosen because it "was capable of exciting those symptoms which make up the fever in a more perfect manner" 18 .
In collaboration with colleagues from the city of São Paulo, we had the opportunity to contact six suspected or confirmed cases of COVID-19. All of them reported anosmia and/or ageusia among other symptoms that we considered characteristic: "breathing was short, almost breathless"; dry and constant cough; cough when speaking (making it difficult to report symptoms over the phone); "cough when you get out of bed and move around".
After identifying the peculiarities of the collective disease COVID-19, we move on to the next step, the selection of the most suitable homeopathic medicine.
According to Hahnemann, "the search for and selection of the homeopathic remedy most suitable in every respect to each morbid state, is an operation which, notwithstanding all the admirable books for facilitating it, still demands the study of the original sources themselves" 19 . The "original sources", mentioned by him, are collections of symptoms (or "Materia Medica") obtained mainly from drug experiments on healthy volunteers, symptoms nowadays denominated as "pathogenetic" 20 .
We searched the original sources using the Materia Medica search function of Vision software 21 . Our findings indicated Thuja occidentalis and Natrum muriaticum as the most homeopathically suitable remedies to the symptoms we had identified. The first hypothesis, Thuja, was given in repeated doses (Thuja LM2) to three adult patients from São Paulo city, by telemedicine. The patients did recover, but the disease seemed to have run its course, with no sign of rapid improvement, as we would be expected in acute cases 22 . We then proceeded to the second hypothesis, Natrum muriaticum (Nat-m).
Pathogenetic symptoms that led us to consider Nat-m as homeopathic to COVID-19 were found (via Vision software) in Hahnemann's "The Chronic Diseases" 23   Although anosmia and ageusia are complaints that took relatively longer to improve, that small sample of patients showed a rapid improvement, reinforcing our hypothesis that Natrum muriaticum might contribute to a faster recovery of COVID-19. To test it, we have decided to run a clinical study.
We chose Nat-m prepared in a fifty-millesimal potency, not only because it is the recommended dynamization by Hahnemann, but also by our experience using LM potencies in randomized controlled clinical trials 26 27 28 29 . Our option for Primary Care stemmed from the possibility of treating milder cases.
SUS Primary Care unities of São Carlos are responsible for screening, treating, and telemonitoring patients with Influenza-like illness during home isolation and referring severer cases to hospitalization. PCR swabs were initially collected in all Primary Care units, later on in regional Primary Care units, but due to logistic screening, testing, treatment, and monitoring are government (SUS) funded.
COVID-Simile study will add on procedures to Primary Care COVID-19 routine.

OBJECTIVE
To investigate the effectiveness and safety of homeopathic medicine Natrum muriaticum (LM2) for mild cases of COVID-19 in Primary Care.

Guidelines
The COVID-Simile study is conducted by the following guidelines: Regulatory  33 .

Participants
Women and men aged 18 years or older, with Influenza-like symptoms and a positive RT-PCR for SARS-CoV-2, will be included. Willingness to give informed consent and to comply with the study procedures is also required. Exclusion criterium: severe acute respiratory syndrome.  34 .

Outcomes
The primary endpoint will be time to recovery, defined as the number of days elapsed before all COVID-19 Influenza-like symptoms 35

Sample size
Considering two independent parallel groups, allocation 1: 1; effect size 0.6; Type 1 error of 5% and Type 2 error of 20% the sample size calculated by G * Power 3.1.9.2 software would be 90 participants. Predicting a 10% follow-up loss, we plan to include 100 participants with COVID-19, randomized to the two treatment arms (Figure 3).

Centralized randomization, allocation concealment, and blinding
In early June, the study Statistician generated a block randomization list, using a 1:1 ratio of the two groups (denoted as A and B) and a web-based tool The study Pharmacist is the only study collaborator who knows the study group (A or B) registered on the card of each envelope, as well as the code, defining whether A or B of the randomization list corresponds to homeopathy or placebo.
The clinical investigators, the statistician, the Primary Care teams, the study collaborators, and the participants will remain blinded from the identity of the two treatment groups until the end of the study. The study Pharmacist has no knowledge or control over which participant each study medication will be sent to.

Monitoring and Telemedicine
According to the COVID- 19  The study Statistician will also have password-controlled access to study website Follow-up pages. If a significant superiority of any of the study's interventions is identified, he will report it to the principal investigator (with no break in blinding). This occurrence will be communicated to CONEP, which will assess the need to adapt or suspend the study.
The clinical investigators have password-controlled access to all website pages.
From follow-up data, they may contact participants to discuss changes in the posology of the study medication (with no break in blinding). The participant will have telephone/WhatsApp access to the principal investigator on study medication issues. If signs of worsening are identified, the principal investigator will contact the respective Primary Care team, suggesting presential medical reassessment, or refer the participant to emergency service, if the worsening happens outside of office hours. Study procedures are illustrated in Figure 4. 1) The study statistician generates a block randomization list of 100 A or B sequences. 2) Executive Coordinator of the UFSCar Health School Unit (UFSCar/USE) prepared 100 A or B cards and places them in 100 sequentially numbered, sealed opaque envelopes, handing them to the study Pharmacist.
3) The Administrative Director of UFSCar/USE decides the study code, i.e., whether A or B will correspond to homeopathy or placebo, and reports that decision on the study Pharmacist. 4) The study Pharmacist opens the envelopes and randomizes each medicine bottle according to the letter written on each card, as well as the code communicated by the Administrative Director. He dispenses the study medication in 100 numbered and sealed packages and hands them to the Administrative Assistant. 5) Primary Care teams treat and monitor mild Influenza-like cases in home isolation. 6) When accessing the test result on the Epidemiological Surveillance website, patients with an RT-PCR positive result receive brief information that a study on homeopathy for Covid-19 is being conducted by professors of UFSCar Medical School. They have the option to get further information by clicking on a button on the webpage if they agree to receive a call from a study researcher. 7) Upon an affirmative answer, the study researcher carries on patients' recruitment, participants enrollment, and monitoring, recording data at the Covid-Simile study website. 8) The Administrative Assistant of UFSCar/USE has password-controlled access to the webpage "Enrollment" of the study website (containing the name, inclusion number, and address of each participant) and sends the Informed Consent (IC) and respective medication package (same Inclusion Number) to the participant's home address, through a delivery company, 9) Which brings back the signed IC to the Administrative Assistant. 10) Clinical investigators monitor participants' progress and can adapt the posology of the study medication by telemedicine, with no break in blinding.
Participants are free to contact the principal investigator on study medication questions.

Data management
Data management will be performed following the ICH-Guidelines for Good Clinical Practice. Participants' data will be inserted or accessed by passwordcontrolled access. Follow-up clinical data will be identified by the participant' Inclusion Number, ensuring confidentiality in data monitoring and analysis. Only the Primary Care health teams, recruitment/telemonitoring researcher, and the principal investigator will have full access to the study website, i.e., to participants' identification and data.

Data analysis
Considering the time to recovery as the main outcome of the study, survival curves will be constructed using the Kaplan-Meier method, for the description of this variable. COVID-Simile Scale scores will be compared between groups by linear mixed-effects models, in which random effects are used to correlate the measurements of the same individual at different times. These models allow the insertion of confounding variables in the comparisons. Comparisons between treatment groups will use the log-rank test. In these analyzes, follow-up losses will be treated as censored information.
Hazard Ratio (HR) will be used as a measure of treatment effect, with its respective 95% confidence interval, estimated by Cox's proportional hazards model. In case the risk proportionality assumption is not met, parametric models will alternatively be used, according to the behavior of the corresponding survival curves. Cox's model and parametric models allow the addition of other control variables for the comparison between treatment groups.
Other secondary measures, expressed as discrete variables, such as the number of hospitalizations, the number of medications will be compared between groups by generalized regression models, based on discrete measures (such as the Poisson distribution).
Outcomes expressed in qualitative variables, such as the occurrence of symptoms and adverse events, will be compared using models based on binary variables (logistic regression).
In all statistical methods, a significance level of 0.05 will be set. As far as we know, this is the only double-blinded, randomized, placebocontrolled trial being conducted in Primary Care to evaluate a homeopathic strategy aiming at reducing the severity and duration of COVID-19 symptoms.
Hypothesizing Natrum muriaticum as able of stimulating COVID-19 "like symptoms" and COVID-19 recovery is hypothesizing effects far beyond the known biological properties of sodium chloride (NaCl), which is a characteristic of the homeopathic pharmacology: inert substances can develop pharmacological effects through dynamizing or potentizing 11 . For instance, salinization has been identified as the main process of soil degradation, but dynamized solutions of sodium chloride (i.e., Natrum muriaticum) can increase the production of healthy seedlings submitted to saline stress 46 . In the emerging field of nanoscience, fifty-millesimal potencies (from LM1-LM30) show "remarkable" presence of nanoparticles of Na and Cl, with contaminants observed in the LM1 degree only 47 . Those nanoparticles may be the source of a low dose of "self-similar information" that will "mobilize neural and cellular defense of the body networks" 48 .
If COVID-Simile results refute the null hypothesis, exactly prepared homeopathic

DISCLOSURE STATEMENT
Cesar AT is co-owner of HN-Cristiano Homeopatia, the pharmacy that has donated the study medication, however, Natrum muriaticum has been in use for over 150 years and is not patentable.

FUNDING SOURCES
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Study medication has been kindly provided by HN-Cristiano Homeopatia -Sao Paulo, SP, Brazil. HN-Cristiano does not participate in the study, except for the donation of study medication. Please give a score (0, 1, 2, or 3)