MD, PhD, FMedSci, FRSB, FRCP, FRCPEd.

 This study aimed to evaluate the effect of Traditional Chinese Medicine (TCM) on patients with gastric cancer following surgery and adjuvant chemotherapy in Taiwan. The cohort sampling data set was obtained from the Registry of Catastrophic Illness Patient Database, a research database of patients with severe illnesses from the National Health Insurance Research Database, Taiwan. Patients who had received a new diagnosis of gastric cancer and had undergone surgery were enrolled. the researchers matched TCM users and nonusers at a ratio of 1 : 3 based on the propensity score, and TCM users were also grouped into short-term and long-term users.

The number of TCM users and nonusers was 1701 and 5103 after applying the propensity score at a ratio of 1 : 3. Short-term users and long-term TCM users were independently associated with a decreased risk of death with HRs of 0.59 (95% confidence interval (CI), 0.55-0.65) and 0.41 (95% CI, 0.36-0.47), respectively, compared with TCM nonusers. The researchers also obtained similar results when they adjusted for covariates in the main model, as well as each of the additional listed covariates. They also observed similar HR trends in short-term users and long-term TCM users among men and women aged <65 years and ≥65 years. The most commonly prescribed single herb and herbal formula in our cohort were Hwang-Chyi (Radix Hedysari; 11.8%) and Xiang-Sha-Liu-Jun-Zi-Tang (15.5%), respectively.

The authors concluded that TCM use was associated with higher survival in patients with gastric cancer after surgery and adjuvant chemotherapy. TCM could be used as a complementary and alternative therapy in patients with gastric cancer after surgery and adjuvant chemotherapy.

This is an interesting study which seems well-done – except for one fatal mistake: even in the title, the authors imply a causal relationship between TCM and survival. Their conclusion has two sentences; the first one speaks correctly of an association. The second, however, not only implies causality but goes much further in suggesting that TCM should be used to prolong the life of patients. Yet, there are, of course, dozens of factors that could interfere with the findings or be the true cause of the observed outcome.

Anyone with a minimum of critical thinking ability should know that CORRELATION IS NOT CAUSATION; sadly, the authors of this study seem to be the exception.

One Response to “Traditional Chinese Medicine Enhances Survival in Patients with Gastric Cancer” – oh really?

  • I don’t know of any adjuvant cancer treatment with an effect of this magnitude. The curves suggest a 30% absolute survival benefit at five years (red curve) and a prolongation of median survival from 2 years to 14 (green curve – the red curve hasn’t got near the median survival yet). The shape of the red curve even suggests that the treatment even protects against death from other causes, such as old age, unless the Chinese cancer population is unusually young. If these effects are real then they will soon be replacing radiotherapy and chemotherapy for many indications worldwide.

    However, before drawing conclusions the first thing to consider when reading any research paper is how the numbers were obtained and what they actually represent. Unfortunately we have rather limited information available here. We also need to consider the type of study. In this instance it is a case-control study, which is generally only ever acceptable for hypothesis generation on which to base prospective studies. The main problem with case-control studies is that by their nature they are always subject to bias, in other words the influence of some important factor which the investigators haven’t considered.

    I would speculate of one possible source of bias. I can imagine that those people who are more likely to use TCM in the first place, perhaps because of their belief in it, or because of poor access to healthcare, would have used this modality at the first onset of symptoms, rather than seeing a gastroenterologist for an endoscopy and a rapid diagnosis. By the time they reached the point of seeking proper medical care, the ones who were still alive and whose tumours were still operable would be the ones with intrinsically less aggressive disease, and therefore the ones who were always going to do better. I would have also thought that they were more likely to be the ones choosing to continue TCM afterwards.

    This is a form of selection bias, which is always problematic when it comes to looking at outcomes. Supposing that you were looking at survival in acute aortic dissection, for instance. You would find that those patients managed in a specialist centre where there are experienced surgeons will always do better than those managed in a district general hospital. Even if the surgery made no difference you would still find this, as it takes time to be referred to the specialist centre and the patients who die before they can get there won’t be included in their survival figures.

    Few people appreciate that the importance of individual biology when it comes to cancer outcomes. A patient whose tumour has the propensity to spread early and grow rapidly is always going to do much, much worse than somebody whose tumour just sits there slowly enlarging. While we talkd about the importance of stage and grade, and early diagnosis, the two biggest factors affecting outcome are the intrinsic nature of the tumour and the overall health of the patient at the time of diagnosis.

    Even fewer people appreciate the importance of understanding the behaviour of numbers in medical research.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Recent Comments

Note that comments can be edited for up to five minutes after they are first submitted but you must tick the box: “Save my name, email, and website in this browser for the next time I comment.”

The most recent comments from all posts can be seen here.

Archives
Categories