Ten TOST Take-Away hints were highlighted in summary:

1. In the survey design phase, start by identifying priority health Results and a concise, manageable set of indicators using standard definitions to facilitate comparability. It is recommended that all of the Rapid CATCH indicators be included. (The purpose of the Rapid CATCH and additional KPC modules is to provide easy access to a core set of standard indicators related to Child Survival and health objectives. These standard indicators are generally comparable to DHS and MICS.)
   
   
2. Don’t get carried away! Repeat, select a concise, manageable set of indictors in addition to the Rapid CATCH. Collect only project data you need and will use. Focus on Results and indicators in which the project may plan  to bring about a measurable change.
   
   Once the results and indicators are chosen, select the questions needed to measure these indicators and construct the questionnaire.
   
   
3. Include the most important background info for factors related to differential child health and nutritional status in your context, e.g. SES, ethnicity, HH Food Security, rural/urban, etc.  Consider if  you are interested in tracking equity in health and nutrition access and coverage between these identified groups.
   
   
4. Pretest, test and then practice, practice, practice with the data collection instrument including weighing infants and children prior to going into the field for actual data collection.
   
   
5. Use a reasonably representative sample of the population you are interested in describing. For cluster sampling, remember to take the design effect into account. For LQAS, respect the principle of random sampling, that each unit (e.g. mother or caretaker of a child under two years) in the population should have an equal and independent chance to be selected in the sample.
   
   
6. Remember that a sample survey provides population estimates.  Don’t forget to report Confidence Intervals and make use of this information when setting targets for measuring change. (Confidence intervals for time 1 and time 2 estimates should not overlap for a change to be considered statistically significant.)
   
   
7. Involve staff and partners in discussing the findings. What do you still need to know about health practices? Can you use qualitative information to increase your knowledge about how and why maternal and child health and nutrition behaviors and practices are what they are?
   
   
8. Use quantitative and qualitative information together as a basis for selecting project results, indicators and targets. Once identified, these results objectives provide the foundation for project planning. The DIP will be developed with the aim of attaining these results targets.
   
   
9. At mid-term (or during periodic monitoring of progress towards results) and final evaluation, look back to the baseline estimates and results targets. Publicly display a graphic representing key results indicator baselines, targets and periodic progress being made towards these targets.
   
   
10. When there is a new project area being added to an existing project area, compare data from where you have been working to the new area where you plan to start intervening. In addition to being able to compare the baseline with the EOP findings, you can now make a cross-sectional comparison between the two areas. This is a much stronger design. Sustainability may also be documented by returning to survey the first project area in conjunction with the EOP survey of the follow-on project in an adjacent area.