by Steve Hickey
Most people understand that a low dose of a drug may be used to prevent a disease. When the disease becomes established, the dose of the drug needs to be increased with the severity of the illness. Antibiotics provide an example: a low oral dose of antibiotic could be used to prevent infection, but if the person becomes ill, it is necessary to increase the dose. In the Cochrane study, the dose-response relationship of vitamin C is apparently considered irrelevant.
Pharmacokinetics
The study of how a drug is absorbed, distributed in the body, and excreted is known as pharmacokinetics. Some substances are poorly absorbed and this may have advantages; for example, magnesium oxide is occasionally used to prevent constipation. Other substances, such as high doses of vitamin C, are rapidly excreted. In understanding the action of a drug or nutrient, it is essential to appreciate the way it is absorbed and excreted.
Drugs and nutrients often interact and bind to protein molecules called receptors, which fit their shape as a key matches a lock. The biological effect depends on the proportion of receptors occupied. If the concentration is low, few receptors are occupied and the response is correspondingly small. As the local drug concentration increases, more is bound to the receptors and the biological effect is greater. At high doses, practically all the receptors are occupied, producing a maximum response. Most drug reactions and many nutrient effects are explained in this way.
Vitamin C interacts with receptors on enzymes and other proteins. However, ascorbate also acts as an antioxidant. When acting as an antioxidant, it donates electrons and the number of electrons available depends on the amount of vitamin C present. The actions of vitamin C are highly dose-dependent and somewhat unusual. For example, low doses of vitamin C are retained in the body and have an effective half-life of somewhere between eight and forty days. By contrast, high doses are excreted many hundreds of times more rapidly. Thus, it is not possible to predict the effects of a large dose using data from a low dose.
We previously mentioned the contraceptive pill as an example of the importance of dose frequency. Contraception usually involves taking a single dose, once a day, through the monthly cycle. The entire month’s supply cannot be given as a single large dose at the start of the month, because such a large dose may have toxic effects and may be ineffective. Moreover, the hormones are expected to be removed from the blood rapidly. In order to be effective as contraceptives, the pill needs to be taken on a daily basis. Women complaining that they became pregnant after missing just one or two pills are describing a process that can be predicted from pharmacokinetics. Missing doses means the blood levels return to baseline in a short interval and the molecular receptors inhibiting pregnancy are no longer occupied. Few doctors would be surprised if women not taking the pill regularly became pregnant.
The excretion half-life of vitamin C is short, so a single tablet will not raise blood levels for more than a few hours. For the rest of the day, blood levels will fall back to the baseline. However, the Cochrane review on vitamin C and the common cold takes no account of the dose interval. Expecting a single, daily gram-level dose of vitamin C to prevent a cold is equivalent to a woman expecting to prevent pregnancy with just one pill a month. The Cochrane review expects vitamin C to work when it is not present in the body. For example, a person taking a gram of vitamin C in the morning may have baseline levels by the afternoon, when she travels home from work and takes her bus ticket from the hand of a cold-infected driver. The virus will enter her body and multiply during the evening and throughout the night. Her plasma vitamin C levels will drop. By the following morning, when she takes another gram of vitamin C, the virus will be well established and growing. The dose may not even raise her blood levels back to baseline and will have little or no effect on the progress of the cold.
The Cochrane Review—“Cargo Cult” Science?
The Cochrane review suggests that massive doses used to treat the common cold can be discounted. Of the reports from Dr. Cathcart and other independent physicians, they say that “their uncontrolled observations do not provide valid evidence of benefit.” Here, we enter a topsy-turvy world, where logic is inadmissible. To the physicians who compiled the Cochrane study, repeated, direct observation and measurement by multiple independent researchers is not only less important than a single clinical trial, it provides no valid evidence at all.
The Cochrane approach indicates that there is apparently no valid evidence that there was a volcanic eruption at Mount St. Helens on May 18, 1980. You may have seen the television record, read it in the newspaper, or perhaps you witnessed it yourself. You may know of scientists who measured the earth tremors, or sampled, measured, and chemically analyzed the rock ejected from the mountain. You may have copies of satellite images that independently recorded and measured the event. However, to the Cochrane Collaboration, these data are irrelevant and provide no valid evidence as they constitute uncontrolled observations.
While Cochrane’s comments suggest that geologists have no valid evidence for the existence of volcanoes or earthquakes, other scientists are in even more trouble. Physics, with its basis of direct, repeatable observation and measurement, should have a new sub-discipline. We suggest the name evidence-based physics, where the only evidence considered valid comes from repeated statistical trials using the methods prescribed by the Cochrane Collaboration. Mathematics and logic may be ignored entirely.
In a summary clearly aimed at the public and press, the Cochrane study states, “Thirty trials involving 11,350 participants suggest that regular ingestion of vitamin C has no effect on common cold incidence in the ordinary population.” This is a general statement implying that vitamin C supplementation, irrespective of dose or frequency, does not prevent the common cold. However, to support such a statement, the authors would need to demonstrate that they had tested the appropriate doses. They have not—the doses tested were far too small. The Vitamin C Foundation’s advice is: “At first sign of cold or flu, begin taking at least 8 grams (8,000 mg) of vitamin C as ascorbic acid every twenty minutes for three to four hours until bowel tolerance. Continue with smaller dosages of 2–4 grams every 4–6 hours for ten days to prevent recurrence.”43
Readers are asked to decide for themselves how much confidence they place in the Cochrane review of vitamin C and the common cold. We agree with Linus Pauling: people should “always be skeptical—always think for yourself.” Efforts to produce evidence-based medicine can be counterproductive if they exclude valuable sources of information. Repeated, easily replicable observation provides, if anything, more direct and reliable evidence than clinical trials.
Tarnished Gold
Conventionally, the core requirement for medical “proof” consists of randomized, placebo-controlled clinical trials, a basic form of experiment for testing therapies on people. Any assertion of the effectiveness of vitamin C tends to be rebutted by a request to see the clinical trial data. As we have seen, all other evidence is effectively discounted. One leading physician even argued that the reports of the remarkable effectiveness of massive doses of vitamin C, by physicians such as Drs. Klenner and Cathcart, could simply be wishful thinking and the placebo effect. The only way to be sure would be to have clinical trials, but such clinical trials, which are fifty years overdue, are not in the works from any conventional scientists in the foreseeable future. Ironically, the medical establishment demands such evidence but withholds the funds that would make it possible—a perfect catch-22.
Clinical trials face public distrust, while the establishment continues to see this form of experiment as the “gold standard” for clinical medicine. Pharmaceutical companies see them as a form of advertising and increasingly dominate trials.44 But the general public are skeptical of such clinical research and the majority of Americans do not trust research information from pharmaceutical companies.45 As time passes, fewer people are willing to accept reported clinical trial results, falling from an estimated 72 percent in 1996 to 30 percent in 2002. Harris Interactive,
a market research and polling firm, estimates that only 14 percent of Americans consider drug companies to be honest, a figure comparable to their opinion of the tobacco, oil, and used car industries.46 Today, 70 percent of people believe that drug companies put profits ahead of patients’ interests.47 This should be expected of commercial companies, as profit is their reason for existence.
Knowledge should always be considered tentative. Even in the face of remarkable observations, such as dying children recovering within minutes of an ascorbate injection,48 we need to be suitably skeptical. In reviewing the available evidence, it is essential to be aware of basic problems with scientific data. For example, some patients simply get better by themselves. Spontaneous remission is more common than people realize and can confound experimental results.49 Demand characteristics are another problem: some patients report what they believe their doctors want to hear. A related effect, effort justification, is when patients feel the need to rationalize the effort and cost of treatment (for example, “This chemotherapy is so tough and expensive, it must be effective!”). However, the factor given most prominence in clinical trials is the placebo. But despite claims by conventional physicians, the placebo effect cannot be used to explain the clinical observations with massive doses of vitamin C.
The Powerful Placebo?
The classic and well-known placebo (Latin for “I shall please”) effect occurs when patients expect to get better and their condition improves, irrespective of the treatment. Doctors assume this effect is so powerful that new drug treatments need to be tested against a dummy pill, in a so-called placebo-controlled trial. These trials are designed to make sure that at least some of the observed improvement is actually caused by the drug rather than the act of treating the patient.
Despite a widespread assumption that the placebo effect is valid, some scientists dispute its existence.50 Numerous factors can cause an apparent placebo effect to arise in experimental data and several other mechanisms might explain placebo effects. Factors such as spontaneous improvement, fluctuation of symptoms, additional treatment, conditional switching of placebo treatment, irrelevant response variables, conditioned answers, neurotic or psychotic misjudgment, and psychosomatic phenomena could contribute to the bias described as the placebo effect. In a well-designed trial, use of a control group is a method to reduce sources of bias.
The placebo effect may be a result of psychological conditioning, a sort of trained response no matter what the treatment. Under this view, the reported effects of vitamin C might be explained as a placebo effect arising from a conditioned response to treatment. An alternative psychological description of the placebo response is the subject-expectancy effect, a bias that occurs when a subject expects a result and unconsciously manipulates an experiment or reports the expectation. This is similar to the observer-expectancy effect, which is when a researcher is expecting a result and unconsciously modifies an experiment, or misinterprets data, in order to find it. Experimenter bias leads to the use of double-blind clinical trials, when neither the experimenter nor the patient knows who is getting the treatment or the placebo until the experiment is concluded and the results are analyzed.
A placebo is an inert pill or substance with no pharmacological effect but which may have psychological therapeutic value. Others extend its meaning to cover any therapy or procedure that has no direct biochemical effect but that may induce a psychological response. The opposite effect, nocebo (Latin for “I shall harm”), is when a patient believes that a treatment will cause harm and experiences adverse side effects when given an inert substance.51 The view of conventional medicine on vitamin C is that any benefit is attributed to the placebo effect, while minor (possibly nocebo) side effects are given great prominence.
The placebo effect is taken for granted in the design of most clinical trials, but the facts are somewhat different. Scientists have used controlled clinical trials to investigate the effectiveness of placebo against no treatment. Numerous trials to investigate the placebo effect have been performed, and an analysis of 130 such experiments was recently reported.52 Strikingly, of the many studies investigated, none was able to distinguish between a placebo effect and the natural course of a disease. It seems that people recovering naturally from a disease can have their improvement attributed to the placebo effect. These experimental results suggest that the placebo effect is severely limited in terms of the physiological responses it can elicit.53 Perhaps unexpectedly, the experimental data indicate that placebo controls are not important in trials of vitamin C with definitive outcomes. A definitive outcome would be the death of the patient, abrupt termination of symptoms, cure of the disease, or some other direct physical response. Sixteen of the 130 investigated trials of the placebo were excluded because they did not have sufficient data on outcomes. Definitive outcomes were present in 32 trials involving a total of 3,795 patients. These trials showed no placebo effect, irrespective of whether the outcome was subjective or objective.
The implications of these findings are clear. While the description of the placebo as a myth or a scam might be going too far, its effects have been grossly overestimated.54 For example, attribution to the placebo effect would be a weak criticism for a study of treatment with vitamin C which produces a five-fold longer survival time in terminal cancer patients.55 Stated simply, if a placebo were such an effective treatment for cancer, it would put current medical practice to shame.
In retrospect, these limitations of the placebo effect should be expected. Basic science and repeatable observation should take precedence over statistical evidence. Nonetheless, to evidence-based medicine, suggesting that a physical event, such as shooting a person through the heart, causes the death of the subject would be speculation. Apparently, it would be necessary to perform a randomized, placebo-controlled clinical trial to show the effectiveness of a physical barrier, such as a Kevlar bulletproof vest. We would need to show statistically that a greater number of people who were shot through the heart died compared with matched controls who wore the vest. Rationally, a bulletproof vest does not need to be compared with a sugar pill in preventing damage from a bullet—we understand the mechanisms involved and the results are definitive.
In the case of vitamin C and viral infections, the claimed effects of complete cessation of symptoms and “cure” are not achievable by any other substance or medication. With the common cold, the disease is generally mild and the discussion somewhat academic. However, modern medicine refuses the patient suffering from severe viral infection, or terminal cancer, the choice of a treatment with vitamin C. Given a disease with no effective alternative therapy and a safe treatment claimed to be highly effective with little risk, the rational decision is clear.
With subjective outcomes, a placebo can have a beneficial effect. Pain is highly subjective and continuous in character and, like psychiatric problems such as depression, might be expected to show a placebo response.56 A continuous outcome is when the effect is scaled from minor to major responses: for example, a patient is asked to estimate the pain they are suffering on a scale of 1 to 10. The placebo effect may be biased toward smaller trials, as it appears to decrease with increasing numbers of patients. A larger number of smaller trials (82), with 4,730 patients in total, had continuous outcomes. The placebo was effective in subjective trials, as might be expected.57 Notably, twenty-seven trials showing benefit from placebo involved the treatment of pain.
The placebo effect has little role in explaining the clinical observations of physicians administering massive doses of vitamin C. If the placebo could be an effective therapy against viral disease, as described for vitamin C by Drs. Klenner and Cathcart, we would not have to worry about such infections. Unfortunately, even the most powerful conventional antiviral medications are largely futile against severe viral diseases.
However, the markedly limited placebo effect is not a reason for minimizing the effect of brain states and psychology in medicine. While the placebo effect is overestimated, the potenti
al of psychological medicine is often underestimated. There is more to psychological medicine than the humble placebo. Nevertheless, it is compelling to compare conventional medicine’s over-ready acceptance of claims for the not-so-powerful placebo with its response to the clinical observations on vitamin C of Drs. Klenner, Cathcart, and Stone.
Random Trials
The randomized, placebo-controlled trial is considered the gold standard of clinical evidence, leading to its cult status in medicine.58 However, without the support of basic science from physiology, pharmacology, and biochemistry, the utility of the clinical trial is severely limited. With little modification, Bradford Hill’s requirements for epidemiology are applicable when considering clinical trials. Clinical trials are merely a technique for measuring the practical value of a therapy, but over-emphasis on the importance of clinical trials devalues information from other sources, such as natural history studies, clinical experience, and case reports. Over the years, evidence for the benefits of vitamin C from these sources has accumulated. While the randomized, controlled trial is important as a practical measure, it is not the only source of scientific information.
Choosing patients for a clinical trial is always a problem. Each person is biologically unique and has an individual response to diseases and treatment.59 Even unconscious bias in choosing patients for an experiment can give invalid results. To reduce the potential for bias, patients in clinical trials are often randomly assigned to the treatment group or the control group. However, effective randomization is difficult to achieve in practice. For example, a suitable criterion is needed for the random selection, such as the order of presentation at the clinic. Even when the groups are fully random, people leaving the study can reintroduce bias. In selective attrition, patients who do not benefit from a treatment or who suffer side effects may drop out, leaving a more positive group of subjects.
Many trials match people in the two randomly selected groups—both groups may be additionally chosen to have a narrow age range or, alternatively, an equal number of males and females in each group. Random selection might otherwise lead, for example, to two sex-biased groups of twelve patients, with four females in one group and ten in the other. Alternatively, a randomly selected treatment group with an average age of seventy-five could have a randomly selected control group of teenagers. These are extreme examples with obvious differences easily noted by the researchers, but since the number of characteristics to match is very large, randomization could easily leave two groups which differ in many unmatched features (blood type, for example). Background knowledge of the underlying disease processes can suggest important characteristics to match in a particular study. Despite this matching, the groups in randomized, controlled clinical trials are substantially different biologically.