OBJECTIVES
ORGANIZATION AND KEY POINTS OF SESSION First Component (15 minutes) VA 's 1-4 Overview of Session In this component stress that the session is directly related to the next session on Preparing Country Presentations. This session requires individual to design an intervention study. Use the first four overheads to review objectives, discuss how you choose strategies and then examine the framework flowchart. When discussing choosing strategies use this as an opportunity to briefly review the many different intervention strategies that could be used that have been covered in the many previous sessions. For the framework point out that the individual stages will be reviewed in the session. You may wish to use this as an opportunity to stress the complementary values of quantitative and qualitative research in the formative studies stage. Stress the importance of listing a variety of intervention studies in Step 3 of the flowchart. When discussing the three possible outcomes in the Follow up stage of the flow chart, point out that negative results are often equally useful. They tell you what interventions not to pursue. Point out the danger of implementing interventions without rigorous testing. Second Component · (10 minutes) VA 's 5-6 Selecting and Intervention These two transparencies stress the two step nature of selecting an intervention. The first step is deciding on a target which requires understanding the problem. This requires characterizing both the key problem behavior and the key target group to change. This requires reviewing the quantitative and qualitative data. In Stage 2 the process of Choosing an intervention requires that a list of possible interventions be produced. Warn participants that for many people the first intervention considered is usually an educational or training intervention. Point out that any list should include a managerial and a regulatory intervention. Briefly discuss the criteria for selecting the intervention which overlap VA 3. You may get asked about whether single or combined interventions should be evaluated. Point out that it is difficult to evaluate the impact of components of a combined intervention and encourage them to sequence such interventions so that you can then assess the relative impact of the different interventions. Third Component · (10 minutes) VA 's 7-9 Study Design Issues In this component you need to briefly review the different types of study design. You need to reject the post only and pre-post study designs as not being worth the effort of doing these studies. When discussing randomized trials and time series studies you need to identify when each can be used. Stress on the unit of analysis issue that the lowest unit is likely to be the facility for a randomized trial. Point out that for hospital studies a times series may be the only possible design. Do highlight the value of having even one other hospital to act as a control in a time series study. You may be asked what do you mean "Multiple measures." The rule of thumb that Steve Soumerai has proposed is at least eight measures before and after the intervention and that 12 measures before and after is better. When discussing VA 8 the three principles may seem obvious but often require discussion. The importance of a comparison group cannot be overstressed. The uses of multiple time measures even in a randomized trial should be explained. Point out that often an intervention has a short term effect and that while one month after an intervention an effect may be noted this may well disappear after 6 months. Stressing the point of focusing on key outcome measures will help with data collection and in writing up the report. Fourth Component · (5 minutes) VA 's 10-11 Sampling Issues This brief component is designed to remind participants about the session on sampling. Point out that sampling is equally important in data gathering for interventions as it is in the original quantitative studies. Point out that all methods may be used and that particularly when substantial travel is required to reach rural facilities clustering may be useful. The issue of sample size is a compromise as always. Suggest that a minimum of 20 facilities need to be included if major changes are expected. Point out that increasing the control group size may be an easier way to increase the total sample size. Fifth Component · (10 minutes) VA 12 Involving Decision Makers This component needs to include participant involvement in the discussion. Who are decision makers? How can they be involved? Point out that if the results are to make a difference the decision makers must accept and buy into the results. This may require the formation of an expert advisory committee, an individual lobbying effort to involve the decision makers or circulating draft proposals for comment. This is also likely to include acknowledging any support or assistance given by any decision makers Sixth Component · (30 minutes) VA 's 13-14 Planning an Intervention These two transparencies take a considerable amount of time to go over. For VA 13 you need to go through each step in detail giving examples and suggestions for choices that have to be made. For each step you may want to ask participants to make suggestions for examples. Point out that all of the steps need to be taken before any work is done. The issue around data analysis needs extensive discussion. Point out that people analyze data not computers. Insist that if data collected cannot be analyzed it should not be collected. Remind participants that statisticians hate to be asked to analyze data after it has been collected. They want to be involved before data collection begins. When discussing presentation of results remember that different people may need differing levels of detail and that some need written materials and others are more impressed by charts and tables. This means that all should be prepared. The involvement of decision makers in the process of presentation of results may be useful. The Gantt chart may take time to explain if participants have not used one before. Work through a simple example of planning a training workshop which participants are likely to be able to identify with. Point out that some activities may be coincident (e.g. seeking permission and preparing materials) but that other activities are sequential (e.g. preparing materials and undertaking training). Point out that with the use of color planned and actual performance can be displayed on the same chart. Seventh Component · (10-30 minutes depending on degree of detail) VA 's 15-16 Budgeting This component can take a lot of time if it is covered thoroughly. The notes in Annex Two are quite thorough and can be used for those who wish to develop detailed budgets. You may also wish to come back to this topic during the session on Preparing Country Presentations. If you going to discuss this section in detail start from the Gantt chart which lists the activities and then ask what resources are required for each activity. Then have them break this down to number of activities e.g. days of work, kilometers travel and then multiply this by the unit cost for each sub activity. Eighth Component · (15-30 minutes depending on how many posters are reviewed) Review of ICIUM Posters Within the session notes there are a number of links to ICIUM Posters. A number of these are useful as teaching aids and are well worth displaying and discussing. The poster Small-group intervention in improving appropriate drug use in acute diarrhoea. Santoso B, Suryawati S, Prawitasari, JE, Ross-Degnan D, is particularly useful to demonstrate. The graphics are very good, and the study shows results from both a randomized study and time series results. A similar study from Indonesia, The impact of problem-based rational drug use training on prescribing practices cost reallocations and savings in primary care facilities, Widyastuti S, Dwiprahasto 1, Andajaningsih, Bakri Z. reports costing data in an interesting fashion. The study from Uganda Prescribing audit with feedback intervention in six regional hospitals and Mulago Referral Teaching Hospital, Uganda. Ogwal-Okeng JW, Anokbonggo WW, Birungi H is one of the few hospital based studies. The results are not dramatic but from a methods point of view this was a well designed study. The study from Gunungkidul is interesting because it demonstrates a long time series with different interventions. It also demonstrates how research can be expanded into programming. The study is Impact evaluation of self-monitoring of drug use indicators in health facilities: experiences from Gunungkidul, Indonesia. Sunartono, Danminto, Suryawati S, Prawitasari J, Bimo, Santoso. The study from Nepal Better primary health care delivery through strengthening the existing supervision/monitoring system. Kafle KK, Pradhan YMS, Shrestha AD, Karkee SB, Das PL, Shrestha N, Prasad RR. Is an interesting study from a methodology point of view. The results are not very impressive but the intervention was relatively weak. The Ugandan study Rational drug use in rural health units of Uganda: effect of national standard treatment guidelines on rational drug use. Kafuko JM, Zirabamuzaale C. Bagenda D. is one of the largest trials of STG’s and supervision. The table describing the design of the study is worth discussing. A community based study that is well worth discussing is Self-learning for self-medication: an alternative to improve the rational use of OTCs, Suryawati S, Santoso B. This poster has some interesting photographs of the community based education occurring. The large two country study A controlled trial of the impact of face-to-face educational outreach on diarrhoea treatment in pharmacies in two developing countries. Ross-Degnan D, Soumerai SB, Goel PK, Bates J, Makhulo J, Dondi N, Sutoto. Adi D, Ferraz-Tabor is well worth reviewing. In addition to the interesting results the methods used could be copied in other settings. Eighth Component · (10 minutes) VA 16-17 Conclusion These final two slides allow you to revisit the issues of selecting methods and the points to consider when developing a strategy which may include multiple different sequenced interventions. Do not spend too long going over the material previously presented but point out that discussion with interested parties is crucial. This leads into the next session Preparing Country Presentations which usually starts immediately after this session. Slide 1
Slide 2
Slide 3
Slide 4
Slide 5
Slide 6
Slide 7
Slide 8
Slide 9
Slide 10
Slide 11
Slide 12
Slide 13
Slide 14
Slide 15
Slide 16
Slide 17
Slide 18
|
