29 Nov 2012


Unlike many of my friends, I do not have a favorite hairdresser. Some shops are too far away. The queue is too long in others. The whole process also takes too much time. Once I asked a hairdresser if I could skip washing and go straight to hair-cutting. He looked at me in disgust (I swear I was clean then). As a result, I keep going from one place to another. Besides, I attend the hairdresser’s infrequently to save time. My hair is always either too short or too long.

Last month, I came across the QB House in Kowloon. They do away hair washing and promise to complete the haircut in 10 minutes. I tried and they really did the job on time. They also did not talk. Exactly what I wanted!

Hairdressers may hate me for not enjoying their service but I guess there are loads of people like me. We just want a simple and quick haircut. If we wanted something else, we would have gone to a beauty salon. Going to work the next morning with a fresh haircut, I could not help but wonder if my patients really want public hospitals to have lobbies fit for a five-star hotel.

22 Nov 2012


Last Saturday, I attended an investigator meeting in Beijing. I did online check-in before the trip and printed the air ticket in my office. To my delight, at the bottom of the ticket was the weather forecast for the next five days. The temperature would be 17 to 20 degrees Celsius. Although I was pleasantly surprised to encounter such warm weather at this time of the year, I duly brought a light jacket only.

As I arrived in Beijing and stepped out of the plane, a chill went down my spine. In fact, a chill went down my entire body. It suddenly dawned on me that 20 degrees Celsius was the temperature of Hong Kong.

15 Nov 2012

Marshmallow Challenge

You may have played a similar version of this game before. Your group is given a bag of uncooked spaghetti, some tapes and strings, and a marshmallow. Your job is to build a tower with spaghetti to support the marshmallow. The group that builds the tallest tower wins.

When Peter Skillman analyzed the results, he noticed something interesting. Somehow, business school graduates consistently performed worse than fresh kindergarten graduates. On average, their towers were 10 inches shorter than those built by children.

What happened? It turns out that business school graduates did what we are all so familiar with – they held meetings. They formulated the best plan. When there were just a few minutes left, they started building their best design. At the last second, the chairman of the group put the marshmallow at the top, and the tower collapsed.

What about the children? They have never heard about meetings, agenda and brainstorming. All kindergarten groups started building right away. Although their towers collapsed time and again, they had plenty of time to try another structure. In the end, more groups would at least have a standing piece.

So, is it at odds with my previous advocation that we should spend much time planning a study? In my opinion, it really depends on whether you have enough information to make decisions. If you do not know anything, pretending you can develop the best plan would not bring you anywhere. This should be time for trial and error and gathering information. Empty talks without execution is, after all, empty talks.

8 Nov 2012


In the face of life and death, the decision on the duration of resuscitation is of secondary importance. In fact, we more often go one step earlier – to decide whether to go for resuscitation in the first place. This decision is closely linked to our ability to predict the prognosis of our patients.

Unfortunately, numerous studies have shown that doctors are poor in prediction. In general, doctors tend to overestimate the survival of their patients.

How about the perception of the patients themselves? In a recent article, Jane Weeks and coworkers studied 1193 patients who received palliative chemotherapy for metastatic lung or colorectal cancer and remained alive 4 months since the diagnosis [N Engl J Med 2012;367:1616-25]. Despite their advanced disease, only 31% of patients with lung cancer and 19% of those with colorectal cancer reported their disease would unlikely be cured by chemotherapy. In contrast, 25% of patients with lung cancer and 36% of those with colorectal cancer thought cure was very likely.

This study is striking in that the majority of the patients had unrealistic expectation about their prognosis. Since Weeks did not study the doctors or the consultation process, it is difficult to determine if the observation resulted from vague explanations by the doctors or patient denial. However, it is noteworthy that all patients in this study chose to receive chemotherapy and survived for at least 4 months. Perhaps it always takes a little optimism to go on in life.

1 Nov 2012


Years ago, a local TV drama pictured a surgeon performing cardiopulmonary resuscitation on a patient who developed cardiac arrest in the middle of an operation. After eight hours of effort, the surgeon refused to let go. “She is already dead,” his colleagues tried to pull him away. “This is human life. We cannot give up,” he yelled back.

Of course, this does not happen in real life. Other than rare occasions such as hypothermia, prolonged resuscitation is futile and does not do the patient any good. However, how long a resuscitation attempt is too long?

In a recent paper, Zachary Goldberger and coworkers reported the findings from an American registry of 64339 patients with cardiac arrests at 435 hospitals [Lancet 2012;380:1473-81]. Overall, 15% of this cohort survived to discharge after successful resuscitation. The average duration of resuscitation was 12 minutes among patients with return of spontaneous circulation, compared to 20 minutes among non-survivors. When centers were classified according to the duration of resuscitation attempts among non-survivors, hospitals in the lowest quartile tried on average 16 minutes and hospitals in the highest quartile tried 25 minutes. The effort was translated into improved outcomes. Patients receiving resuscitation in hospitals in the highest quartile of attempt duration were 12% more likely to survive to discharge.

Thus, it is reassuring to learn that attempting more does help. This, however, still does not answer the original question – How long should we try then?