(1995; 68 pages)
1. Databases on acupuncture have a cultural basis but form the essential first step of any research project. Learning from previous work is an inherent part of the scientific process and that database can provide suitable reference for published work.
2. Descriptive research outlines the observed and uncontrolled effects of acupuncture with respect to:
• traditional Chinese medicine and its variations;
• the cultural aspects of each country’s medical system;
• the process or technique of the acupuncture utilized; and
• outcomes (objective and subjective).
A descriptive study could be used as the basis for more detailed investigations.
3. Randomized controlled clinical trials. Related problems and difficulties are outlined elsewhere.
New research strategies based on a realistic assessment of the cost and the cultural and political environments in which health care operates need consideration. These include:
a. pragmatic research that compares outcomes from patients receiving different treatment “packages” (conventional and traditional); and
b. developmental research that allows us to develop a better understanding of cost and cost-effectiveness.
Cohort studies are essentially uncontrolled prospective investigations in which detailed data are kept and then analysed to evaluate the effects of acupuncture. The advantages of cohort studies are that they allow the researcher to formulate a coherent database upon which more detailed clinical trials can be conducted. However, too often, the protocol for these studies is poorly formulated and the data collection is incomplete and inadequate. They represent an important first step in a multimodal research approach to acupuncture. Nevertheless, the conclusions generated by such evaluation must be treated with caution until they can be confirmed with further appropriate research. For instance, they can provide information about which types of patients are likely to respond best to acupuncture for a particular condition. This helps a researcher decide on the criteria for entry into an RCT, but cohort studies, however rigorously constructed, do not prove the value of acupuncture.
Retrospective studies/case control studies
For the purpose of this chapter, retrospective studies will refer to retrospective observations limited to a relatively small number of patients.
Retrospective research is valuable in that it may provide preliminary data on the effectiveness of a particular treatment. The difficulties usually encountered relate to the fact that often relevant data have not been collected consistently and are therefore not available to allow for proper statistical analysis. Also, proper control groups are usually not available although the limitation may be partially compensated by the use of historical matched controls. In addition, the small number of observations may reflect spurious effects rather than generalizable phenomena. The most common retrospective studies are the case control studies in which possible matching of patients and controls is done according to outcome.
Sequential trial design
Sequential trial design does not fix the sample size beforehand and the trial is conducted based on the comparison of two groups. Generally, sequential trial can be completed with a minimum of patients consistent with a statistically significant result, but unfortunately, a sequential trial can only be employed in certain situations.
In sequential trial designs, it is difficult to allow for more than one response variable or for more than two treatments and it will be administratively complex if the trial is multicentre. Sequential trial designs are likely to be of limited use to those therapies where treatment results are generally known too late to limit patient entry.
In commonly used sequential trials, patients enter in matched pairs, one member in each pair receiving (at random) the treatment to be tested and the other a placebo (or alternative treatment). Success or failure of the treatment is determined on each pair of patients sequentially as soon as the results become available and eventually, for each pair, when both treatments are a success or both treatment are a failure, they are discarded from analysis. Usually, a score of +1 is given to an outcome in which tested treatment is a success and placebo alternative treatment a failure, and a score of -1 to an outcome in which placebo or alternative treatment is a success and tested treatment a failure. As the trial proceeds, a cumulative score is kept. It is evident that if tested treatment is markedly superior to alternative treatment, an increasing positive score will be accumulated, whilst an increasing negative score will accumulate in the reverse case. A sequential analysis chart is usually used in the analysis of a clinical trial.
Single subject experimental designs
Single subject experimental designs (single case designs, n of 1 trial) were developed in the field of psychology and have recently been adapted for clinical research.
Single case designs can evaluate the effectiveness of various specialized acupuncture methods in patients with a variety of individual differences. They are easy to adopt as an exploratory study and their cost is relatively low. Various single subject experimental designs are proposed for clinical trials. In this section, two simple designs are introduced.
A reverse design (AB method) is the simplest n of 1 trial in which baseline data (A) are collected and their stability confirmed before treatment. Then a specific treatment is applied and evaluated by the practitioner. Use of a time series analysis is recommended. Repetitive measurements (ABABAB...) increase the plausibility of results.
In an alternating design, different treatments are repetitively applied in a random order, then the data can be analysed using conventional statistical methods.
However, the latter two techniques are obviously not indicated in the case of acupuncture treatments that have long-lasting or irreversible effects. The results of single subject experimental designs cannot be easily generalized; however, the usefulness of the single subject experimental designs in clinical research on acupuncture should be noted.
Clinical audit is a process through which the management of a patient may be improved. The audit cycle is an expansion of a critical approach to the clinical management of patients. Comprehensive data on each patient are required. The aim of the audit is to provide the “best” treatment for a particular patient or illness by continually evaluating the treatment technique against treatment outcome. It is usual to discuss such information within a group of clinicians so that the audit cycle of treatment, critical appraisal and then an improved treatment regimen can be continually developed. The process of a clinical audit creates a positive and supportive environment among acupuncturists. This environment is essential for research development and also allows for the evolution of a research culture and the development of good practice guidelines within acupuncture. The development of “best practice” facilitates the approach required for other research techniques such as RCTs and also acts directly to benefit the patient.
In the drug evaluation field it has been recognized that information obtained from premarketing clinical trials (phase I, II, III) is imperfect because:
1. the number of patients are limited in the premarketing stage;
2. the drug is used in a range of different conditions after marketing and combined with the other drugs and treatments in complex clinic situations.
Thus, a mechanism called postmarketing surveillance (PMS) has been developed to collect and analyse information in non-trial settings, i.e. general clinical use at primary care level. Originally PMS was designed to collect information about safety but it gradually began to cover effectiveness.
“Pharmaco-epidemiology” is the term used to describe this field. It covers reporting systems, statistical analysis and the necessary drug regulations which allow outcome information to be obtained.
This method can be used in clinic research in countries where acupuncture has legal recognition and in those where it may be recognized in the near future. This methodology could be called “acupuncture epidemiology”. The unregulated status of acupuncture in some countries is an obstacle to the development of this area because people who use acupuncture will be reluctant to participate. The official recognition of acupuncture is a prerequisite for the development of acupuncture epidemiology.
“Outcomes research” in relation to acupuncture is synonymous with acupuncture epidemiology. In some countries, information technology can be utilized; computerized databases of health information which cover every aspect of health are a potential source for this research. Medical cards which store all health information for an individual could also be used. Outcomes relate not only to safety, but also, its effectiveness and economic value, i.e. cost-effectiveness. Cohort studies are prospective descriptive studies and can also be utilized in the context of acupuncture epidemiology.
Anthropological research requires an understanding of the social and cultural environment in which acupuncture is practised. This may directly impact on clinical research as it may explain the reasons that, in some countries, there are cultural difficulties in the development of controlled clinical trials and in obtaining informed consent It also involves cooperation from social scientists and, thus, will allow nongovernmental organizations (NGOs) and government organizations to understand their country’s health service requirements and the demands of the population. The socioeconomic and sociopolitical importance of this research is obvious and it must go hand in hand with clinical trials.