The 2003 WHO report concluded that the benefits of acupuncture were either 'proven' or 'had been shown' in the treatment of ninety-one conditions. It was mildly positive or equivocal about a further sixteen conditions. And the report did not exclude the use of acupuncture for any conditions. The WHO had given acupuncture a ringing endorsement, reinforcing their 1979 report.

It would be natural to assume that this was the final word in the debate over acupuncture, because the WHO is an international authority on medical issues. It would seem that acupuncture had shown itself to be a powerful medical therapy. In fact, the situation is not so clear cut. (    ア    ), as we shall see, the 2003 WHO report was shockingly misleading.

The WHO had made (A)two major errors in the way that it had judged the effectiveness of acupuncture. The first error was that they had taken into consideration the results from too many trials. This seems like a perverse criticism, because it is generally considered good to base a conclusion on lots of results from lots of trials involving lots of patients-the more the merrier. If, however, (    イ    ), then those particular results will be misleading and may distort the conclusion. Hence, (    ウ    ) had it implemented a certain level of quality control, such as including only the most rigorous acupuncture trials. Instead, the WHO had taken into consideration almost every trial ever conducted, because it had set a relatively low quality threshold. Therefore, (    エ    ).

The second error was that the WHO had taken into consideration the results of a large number of acupuncture trials originating from China, whereas it. would have been better to have excluded them. At first sight, this rejection of Chinese trials might seem unfair and discriminatory, but there is a great deal of suspicion surrounding acupuncture research in China. For example, let's look at acupuncture in the treatment of addiction. Results from Western trials of acupuncture include a mixture of mildly positive, equivocal or negative results, with the overall result being negative on balance. By contrast, (    オ    ). This does not make sense, because (    カ    ). Therefore, (    キ    ). The crude reason for blaming Chinese researchers for the discrepancy is that their results are simply too good to be true. This criticism has been confirmed by careful statistical analyses of all the Chinese results, which demonstrate beyond all reasonable doubt that Chinese researchers are guilty of so-called (B)publication bias.

Before explaining the meaning of publication bias, It is important to stress that this is not necessarily a form of deliberate fraud, because it is easy to conceive of situations when it can occur due to an unconscious pressure to get a particular result. Imagine a Chinese researcher who conducts an acupuncture trial and achieves a positive result. Acupuncture is a major source of prestige for China, so the researcher quickly and proudly publishes his positive result in a journal. He may even be promoted for his work. A year later he conducts a second similar trial, but on this occasion the result is negative, which is obviously disappointing. The key point is that this second piece of research might never be published for a whole range of possible reasons: maybe the researcher does not see it as a priority, or he thinks that nobody will be interested in reading about a negative result, or he persuades himself that this second trial must have been badly conducted, or he feels that this latest result would offend his peers. Whatever the reason, the researcher ends up having published the positive results of the first trial, while leaving the negative results of the second trial buried in a drawer. This is publication bias.

When this sort of phenomenon is multiplied across China, then we have dozens of (    ク    ) positive trials, and dozens of (    ケ    ) negative trials. Therefore, when the WHO conducted a review of the (    コ    ) literature that relied heavily on Chinese research its conclusion was bound to be skewed-such a review could never take into account the (    サ    ) negative trials.

注
• acupuncture: はり治療
• equivocal: はっきりしない
• ringing endorsement: 明確な是認
• reinforce: 強化する
• perverse: ひねくれた
• distort: ゆがめる
• rigorous: 厳格な
• threshold: 閾値、最低基準
• discriminatory: 差別的
• addiction: 中毒、依存症
• discrepancy: くい違い
• fraud: 詐欺行為
• skew: ゆがめる

• 問1. 空所( ア )に入る表現を(a)～(f)の中から1つ選び、その記号を答えなさい。
  • (a) As a rule
  • (b) Continuously
  • (c) For instance
  • (d)　Fortunately
  • (e) Regrettably
  • (f) Sooner or later
• 問2. 空所( イ )～( エ )にはそれぞれ次の3つのいずれかが入る。各空所に入るものの記号を答えなさい。
  • (a) some of the trials have been badly conducted
  • (b) the final report was heavily influenced by untrustworthy evidence
  • (c) the sort of overview that the WHO was trying to gain would have been more reliable
• 問3. 空所( オ )～( キ )にはそれぞれ次の3つのいずれかが入る。各空所に入るものの記号を答えなさい。
  • (a) Chinese trials examining the same intervention always give positive results
  • (b) either Eastern researchers or Western researchers must be wrong-as it happens, there are good reasons to believe that the problem lies in the East
  • (c) the efficacy of acupuncture should not depend on whether it is being offered in the Eastern or Western hemisphere （注 efficacy: 効力）
• 問4. 空所( ク )～( サ )にはいずれも'published'か'unpublished'のいずれかが入る。'published'が入る空所名をすべて丸で囲みなさい。
• 問5. WHOが2003年に発表した報告書に関して適切なものを次の中から1つ選び、その記号を答えなさい。
  • (a) 1979年の報告書の結論をくつがえして、はり治療が多くの症状に有効であることを認めた。
  • (b) 1979年の報告書にも述べられていたはり治療に対する懐疑的な態度を、より明確な否定にまで強めた。
  • (c) 従来の多くの研究を検討した結果、はり治療の有効性は広範に認められるという肯定的な評価を下した。
  • (d) はり治療の効果を強調しすぎる傾向に対して、有効ではないケースがあることを示して注意を喚起した。
  • (e) はり治療はどのような症状に対して有効であり、どのような症状に対しては有効ではないかを示した。
• 問6. 下線部Aの「2つの間違い」とは何か。次の中から2つ選び、その記号を答えなさい。
  • (a) 行われた研究の質を十分考慮することなく、ただ数を多く集めて結論を出してしまったこと。
  • (b) 肯定的な結果だけを重視して、否定的な結果を軽視してしまったこと。
  • (c) 欧米で行われた研究を重視して、中国での研究を十分に考慮しなかったこと。
  • (d) 中国で行われた研究を欧米で行われた研究と平等に扱ってしまったこと。
  • (e) わずかな数の研究の結果から性急に一般的な結論を導き出してしまったこと。
• 問7. 下線部Bのpublication biasとはなにか。次の中から適切なものを1つ選び、その記号を答えなさい。
  • (a) 同じ内容の論文でも、有力な雑誌に発表されると高く評価され、無名の雑誌に掲載されると無視されてしまう傾向。
  • (b) 肯定的な結果が出た場合には論文として発表されるが、否定的な結果が得られた場合には発表されずに終わってしまうという傾向。
  • (c) 肯定的な内容の発表は注目されるが、否定的な内容の発表は無視されがちであるという傾向。
  • (d) 最初に行った研究の結果は発表するが、確認のために二度目に行った研究の結果は発表されないことが多いという傾向。
  • (e) 雑誌の編集者の考えが影響して、掲載される論文の内容に偏りが生じてしまうという傾向。