Valkee's most important marketing claim:
Did a study really show that the earlight device cures 92% of severely winter-depressed patients?

Background:
Clinical trials in depression measure, in the first place, the placebo effect.60 Known for decades to experts as "industry's dirty little secret",61 this fact is brought up by the lay media from time to time as news.62 63 Placebo can have dramatic effects in such trials, often producing the same benefit as active treatment.64 This placebo response has been shown in more than 1000 trials,65 and dozens of high-profile meta-analyses, making it the most basic factor to consider when looking at results.

All unspecific treatment effects are summarized as placebo response. It is always present, while a tested treatment itself may work or not - a quack will never have zero points. It can neither be predicted nor calculated and differs enormously from study to study.66 Therefore, comparisons between active treatments across trials are invalid, and efficacy can only be demonstrated with a placebo or active control. 7 

Many factors are known to influence the placebo effect: In controlled studies, for instance, the information to get some active treatment is boosting the placebo response by 15%points.67 Probably most influential are a patient's visits at the investigating doctor's office, each increasing the placebo effect by 0.8 HAMD points in a linear manner - the more visits, the better the feeling.68 Naturally, companies seek to minimize the placebo response in efficacy studies, to make the working treatment look better.

Valkee did the opposite and designed a marketing trial with a maximum placebo effect. They set up an open study, where Valkee told all participants they would get a brand new, state-of-the-art treatment. With visits to the clinic every workday for 4 weeks, employing a highly technical, impressive procedure, "cure" rates of close to 100% can be achieved without a really working treatment.

How many patients? (Screenshot from Valkee.com, 2011)

How they did it:
13 persons were carefully selected as participants, using criteria known to exclude the vast majority of real-world patients.69 Demonstrative earlight treatment was applied on 5 days a week for 4 weeks. One visit per week held also a rating of depressive symptoms by the investigator, using the HAMD scale. Other scales, including the BDI self-rating questionary, were used to collect supportive information.

At study start, patients rated themselves on the BDI with an average of 15 points. This is the equivalent of a mild depression,79 which, according to international guidelines, does not require any specific treatment.78 A person with such a BDI would not even be suspected to have any clinically significant depression, because the diagnosis threshold in Finland is at least 17 points, even in screening.77 After 4 weeks of treatment, in 12 of 13 persons (92%) the values were so low, that those participants were counted as cured from their minimal symptoms.

Valkee's investigators instead rated the patients at study start with an average of 23 points on the HAMD scale (severe depression). This extreme difference between perceived symptoms and investigator rating would have needed explanation before publication in any quality journal, since such "score inflation" is known to bias trial results massively.99 The score then fell to less than 8 (remission) in 10 out of 13 patients: 77% were "cured".

What does it mean:
An unimpressive number, considering that true bright light therapy has been shown to cure as much patients in only 10-14 days in open studies.70 From a scientific viewpoint, this is an inconsequential result with low credibility for some outsider treatment never heard of before:

How was it published:
Publication in a peer-reviewed journal is a basic scientific standard, meaning that independent reviewers check the worthiness of one's work before accepting it for print. Valkee's pseudoscience did not pass this quality check in any such journal. They found a place to slip through in Medical Hypotheses, where authors themselves name the reviewers.71 A place for everything rejected by serious journals, ranging from L.Ron Hubbards detox method,72 to twinning Asians and Down persons,73 or treatment of nasal congestion by masturbation.74

Valkee's co-founder Juuso Nissilä blamed for this failure a conspiracy of journal editors and industry against his revolutionary findings.75

What was the impact:
The "controversial" article was completely ignored by the scientific community. In 1,5 years since it's publication it was not cited by any publication, except the Valkee group themselves, and a polish website.76 In Finland, the lay media absurdly hailed this pseudoscience as "finnish top research", transporting Valkee's marketing message into every household.

Conclusion:
In a small low-quality study designed for a maximum placebo effect, Valkee managed to get a "cure" rate of 77%. The study setting alone explains this number. Nothing points to a treatment effect of the earlight device. Not adhering to basic scientific standards, this pseudoscience was exclusively done, and used, to market the quack product.

Claims that 9 of ten winter depression sufferers were cured by Valkee's device are false. Using these numbers without telling how they were produced is lying by omission.

See also: The placebo-controlled trial, were a more realistic design brought "cure" rates so low that Valkee did not report them.

References & Sources:

7 The European Agency for the Evaluation of Medicinal Products: Note for Guidance on Clinical investigation of medicinal products in the treatment of depression. 25.4.2002, retrieved 6.2.2013.

60 Mora MS, Nestoriuc Y, Rief W: Lessons learned from placebo groups in antidepressant trials. Philos Trans R Soc Lond B Biol Sci 2011;366(1572):1879-88.

61 Hollon SD, DeRubeis RJ, et al: The emperor's new drugs: Effect size and moderation effects. Prevention & Treatment 5(1) 2002, Art. 28.

62 Boseley S: Prozac, used by 40m people, does not work say scientists. The Guardian, Feb 26 2008.

63 MTV3 Uutiset: Psykologi: Suurin osa depressiolääkkeiden tehosta perustuu lumevaikutukseen. MTV3 7:n uutiset, website 9.7.2012.

64 Kirsch I, Deacon BJ, et al: Initial severity and antidepressant benefits: a meta-analysis of data submitted to the Food and Drug Administration. PLoS Med 5(2) 2008: e45.

65 Ioannidis John PA: Effectiveness of antidepressants: an evidence myth constructed from a thousand randomized trials? Philosophy, Ethics, and Humanities in Medicine 2008,3:14.

66 Walsh BT, Seidman SN, et al: Placebo response in studies of major depression: variable, substantial, and growing. JAMA 2002;287(14):1840-7.

67 Enck P, Klosterhalfen S, et al: The placebo response in clinical trials: more questions than answers. Philos Trans R Soc Lond B Biol Sci 2011;366(1572):1889-95.

68 Posternak MA, Zimmerman M: Therapeutic effect of follow-up assessments on antidepressant and placebo response rates in antidepressant efficacy trials. Metaanalysis. British Journal of Psychiatry (2007) 190: 287-292.

69 Zetin M, Hoepner CT: Relevance of exclusion criteria in antidepressant clinical trials: a replication study. J Clin Psychopharmacol 2007;27(3):295-301.

70 75% in the pooled group, see Terman JS, Terman M, et al: Efficacy of brief, intense light exposure for treatment of Winter depression. Psychopharmacol Bull 1990 26(1):3-11.

71 Elsevier: Medical Hypotheses: Guide for Authors. Retrieved on Jul 3 2013.

72 Cecchini M, LoPresti V: Drug residues store in the body following cessation of use: impacts on neuroendocrine balance and behavior - use of the Hubbard sauna regimen to remove toxins and restore health. Med Hypotheses 2007;68(4):868-79.

73 Mafrica F, Fodale V: Down subjects and Oriental population share several specific attitudes and characteristics. Med Hypotheses 2007;69(2):438-40.

74 Zarrintan S: Ejaculation as a potential treatment of nasal congestion in mature males. Med Hypotheses 2008;71(2):308.

75 Nissilä J: "Akuutti", 19:30 on 17 Jan 2012, channel YLE 2.

76 Google scholar search for the article "Can transcranial brain-targeted bright light treatment via ear canals be effective
in relieving symptoms in seasonal affective disorder? – A pilot study", Jul 4 2013.

77 Nuevo R, Lehtinen V, et al: Usefulness of the Beck Depression Inventory as a screening method for depression among the general population of Finland. Scand J Public Health 2009;37(1):28-34.

78 National Institute for Health and Clinical Excellence (NICE): Clinical Guideline 90: Depression in adults. Issued October 2009, p.9.

79 Wikipedia: Beck Depression Inventory. Accessed Sep 21, 2013.

99 Landin R, DeBrota DJ, et al: The impact of restrictive entry criterion during the placebo lead-in period. Biometrics. 2000 Mar;56(1):271-8.

Antidepressant studies are cited here for universal features, like patient characteristics and trial design. Valkee uses mainly pharmaceutical industry's methods in research and marketing.