Search

This concluding module attempts to synthesize the key learning points discussed during the course of the previous ten sets of modules on methodology and biostatistics. The objective of this module is to discuss how to present a model research proposal, based on whatever was discussed in the preceding modules. The lynchpin of a research proposal is the protocol, and the key component of a protocol is the study design. However, one must not neglect the other areas, be it the project summary through which one catches the eyes of the reviewer of the proposal, or the background and the literature review, or the aims and objectives of the study. Two critical areas in the "methods" section that cannot be emphasized more are the sampling strategy and a formal estimation of sample size. Without a legitimate sample size, none of the conclusions based on the statistical analysis would be valid. Finally, the ethical parameters of the study should be well understood by the researchers, and that should get reflected in the proposal. [ABSTRACT FROM AUTHOR]

As researchers, we often collect data on a clinical record form or a questionnaire. It is an important part of study design. If the questionnaire is not well designed, the data collected will not be useful. In this section of the module, we have discussed some practical aspects of designing a questionnaire. It is useful to make a list of all the variables that will be assessed in the study before preparing the questionnaire. The researcher should review all the existing questionnaires. It may be efficient to use an existing standardized questionnaire or scale. Many of these scales are freely available and may be used with an appropriate reference. However, some may be under copyright protection and permissions may be required to use the same questionnaire. While designing their own questionnaire, researchers may use open- or close-ended questions. It is important to design the responses appropriately as the format of responses will influence the analysis. Sometimes, one can collect the same information in multiple ways - continuous or categorical response. Besides these, the researcher can also use visual analog scales or Likert's scale in the questionnaire. Some practical take-home points are: (1) Use specific language while framing the questions; (2) write detailed instructions in the questionnaire; (3) use mutually exclusive response categories; (4) use skip patterns; (5) avoid double-barreled questions; and (6) anchor the time period if required. [ABSTRACT FROM AUTHOR]

In this module, we have discussed study designs that have not been covered in the previous modules -- ecologic studies and natural experiments. In an ecologic study, the unit of analysis is a group or aggregate rather than the individual. It may be the characteristics of districts, states, or countries. For example, per capita income across countries, income quintiles across districts, and proportion of college graduates in states. If the data already exist (such as global measures and prevalence of diseases, data sets such as the National Family Health Survey, census data), then ecologic studies are cheap and data are easy to collect. However, one needs to be aware of the "ecologic fallacy." The researcher should not interpret ecologic level results at the individual level. In "natural experiments," the researcher does not assign the exposure (as is the case in interventional studies) to the groups in the study. The exposure is assigned by a natural process. This may be due to existing policies or services (example, one city has laws against specific vehicles and the other city does not); changes in services or policies; or introduction of new laws (such helmet for bikers and seat-belts for cars). We would like to encourage researchers to explore the possibility of using these study designs to conduct studies. [ABSTRACT FROM AUTHOR]

In a clinical trial, study participants are (usually) divided into two groups. One group is then given the intervention and the other group is not given the intervention (or may be given some existing standard of care). We compare the outcomes in these groups and assess the role of intervention. Some of the trial designs are (1) parallel study design, (2) cross-over design, (3) factorial design, and (4) withdrawal group design. The trials can also be classified according to the stage of the trial (Phase I, II, III, and IV) or the nature of the trial (efficacy vs. effectiveness trials, superiority vs. equivalence trials). Randomization is one of the procedures by which we allocate different interventions to the groups. It ensures that all the included participants have a specified probability of being allocated to either of the groups in the intervention study. If participants and the investigator know about the allocation of the intervention, then it is called an "open trial." However, many of the trials are not open - they are blinded. Blinding is useful to minimize bias in clinical trials. The researcher should familiarize themselves with the CONSORT statement and the appropriate Clinical Trials Registry of India. [ABSTRACT FROM AUTHOR]