CDC Module 2 Lesson 4

pdf

School

Texas A&M University *

*We aren’t endorsed by this school

Course

615

Subject

Statistics

Date

Apr 3, 2024

Type

pdf

Pages

Uploaded by MagistrateEnergyHerring41

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO L ESSON 4: P HASES 3 AND 4 – D ATA E NTRY , A NALYSIS , AND W RITING THE R EPORT I NTRODUCTION In lesson 3, you learned how to define the scope and nature of the questionnaire and how to design the questionnaire. Now we will cover phases 3 and 4, Data Entry and Analysis , and Writing the Report . We will discuss the process for analyzing data collected from a questionnaire using a statistical software package and learn how to use a preliminary and final reporting structure to report the results from the questionnaire and the data analysis. D ATA E NTRY AND A NALYSIS You will use statistical software packages to analyze data collected from both paper and electronic questionnaires. Although a variety of statistical software packages are available, we recommend Epi Info™ software: it is user-friendly and free of charge. Any statistical software package is acceptable as long as it allows you to “weight” data (e.g., SAS). Weighting data is a key requirement for any software package considered for an RNA data analysis. Data Entry To ensure the quality of data collected, whether electronically or paper-based, consider how you will enter and analyze the data. Table 9 details considerations for entering and analyzing data. Table 9. Considerations for data entry. 26 Paper Form Electronic Device • Enter into an electronic dataset using software that is familiar, easily accessible, and maintainable • Build data entry platform before conducting fieldwork • Provide training for data entry staff to enter data accurately • Merge collected data into a single file that statistical software can analyze (e.g., Epi Info™, SAS)

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO Data Analysis For tutorials in creating the database, entering data, and running analysis, please visit the CDC’s Epi Info™ website at http://wwwn.cdc.gov/epiinfo/user-guide/ and follow the User Guide To adjust the data to account for a complex sampling design, such as the two-stage cluster sampling methodology, weighted frequencies are used to represent the entire target population (i.e., the sampling frame). Note that in most cases, nonweighted analyses will not represent the entire target population and may bias estimates. Remember, households selected in cluster sampling have an unequal probability of selection. To avoid biased estimates, data analyses should include a mathematical weight for selection probability. Weighting involves assigning a weight for each household for which an interview is completed. The weight is based on the household’s probability of selection. Once all data are merged into a single electronic dataset, a weight variable is added to each surveyed household by use of the formula in Figure 4 (see DEMO data) 26 ibid Figure 4. Weight variable formula Total number of households in the sampling f rame Weight = (number of households interviewed in cluster) * (number of clusters selected)

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO The numerator is the total number of households in the sampling frame; that number will be the same for every assessed household. Remember, the sampling frame is defined as the entire assessment area in which the RNA is conducted. In only one instance will all the weights be equal. If you complete seven interviews in each cluster, then each household will receive the same weight – assigning weighting will not be required. For example, if information was obtained from exactly seven households in 30 clusters, the denominator for every household would be 7 (number of interviews) * 30 (number of clusters selected) which equals 210. In this case, the sample itself was weighted because all households in the sample had an equal probability of being selected. This outcome is possible, but unlikely to occur. The more likely scenario is to have households where people are not home, households refuse to participate, or insufficient households in the clusters. When this occurs, the denominator will be different for each surveyed household depending on the cluster from which the household was selected. Households from the same cluster will have the same weight, but weights will differ between clusters. For example, if only five completed interviews occurred in a cluster, the denominator of the weight for each of the five surveyed households would be 5 (number of interviews) * 30 (cluster size) = 150. Example of Applying Weights: India Tsunami An RNA conducted in India after a major tsunami will illustrate the process of applying weights using sampling data collected from an RNA. In stage one of sampling, 30 clusters were selected representing a total of 19.370 households. The goal was to conduct 210 interviews, but only 187 were completed. Figure 5 provides a sample dataset showing the number of interviews per cluster and the assigned weight for each household interviewed. For the purpose of calculating the “weight” column (highlighted in yellow), an additional column was added, “”# interviews”, to represent the number of households interviewed within the clusters (highlighted in blue). Note that the weighting formula circled in red is the same calculation with the three variables introduced above.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO Figure 5. Sample dataset showing the number of interviews per cluster and the assigned weight for each house interviewed. 27 27 ibid Calculating Weighted Frequencies After assigning weights, the next step was to calculate the frequencies for each of the interview questions. For the purposes of this RNA, EPI Info ™ was the statistical software. To calculate weighted frequencies in Epi Info™ “classic mode,” the following steps were performed (use DEMO data): • Step 1: Import the data file with the weight just created • Step 2: Click on “Frequencies” along the left hand column • Step 3: In the “Frequency of” box, select variable(s) for which you would like results • Step 4: In the “Weight” box, select the variable ”WEIGHT” that was just created • Step 5: Click “OK” (see Figure 6) and generate a report providing the estimates. Figure 6. Frequency analysis window showing selected variables and weight

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO Figure 7 displays the Epi Info™ output window with the selected variables from the previous step, followed by a table for each selection. Save the output tables and use them in the final report. Figure 7. Example of Epi Info™ output window showing weighted frequencies Nonweighted frequency To obtain nonweighted estimates, follow the above instructions but do not assign a variable in the “weight” box. Applying weights provides projected estimates that can be generalized to every household in the assessment area or sampling frame. The unweighted frequency only shows results from your RNA sample. Table 10 shows the nonweighted and weighted frequencies for a specific question from a Tsunami RNA. Table 10. Nonweighted and weighted frequencies of current source of electricity following a tsunami 28 Sources of Electricity Indicators Sources of Information Frequency % Frequency % 95% CI Power company 137 74.1 14,190 74.0 61.9–86.0 Gasoline generator 15.7 74.1 3,200 16.7 7.6–25.7 None 19 10.3 1,789 9.3 3.8–14.8 28 ibid

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO Remember that weighted analysis does not account for changes that may occur in the number of households between the time of the census and the time of the assessment (e.g., the number of households per cluster may have changed between when the census was conducted and when the RNA was conducted). Thus, despite attempts to arrive at unbiased estimates, the reported frequencies may lack precision. Calculation of 95% Confidence Intervals Once the weighting has been established, you will need to calculate Confidence Intervals (CIs) . When conducting weighted estimates, the RNA team should use 95% CIs. CIs represent the reliability of the weighted estimate. Do the following steps to calculate a 95% confidence interval in Epi Info™ “classic mode,” (see DEMO data): • Step 1: Import the data file into Epi Info™ • Step 2: Select “Complex Sample Frequencies” command under “Advanced Statistics” on the left hand side. • Step 3: In the “Frequency of” box, select variable for which you interested • Step 4: In the “Weight” box, select the “WEIGHT” variable that was created • Step 5: in the “Primary Sampling Unit” box, select the “Cluster Number” variable • Step 6: Click “OK” (see Figure 9) to generate a table providing confidence intervals Figure 8. Selected variable for calculation of complex sample frequencies

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

Figure 9. Example of Epi Info™ output window showing complex sample frequencies D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO Response Rates Calculating response rates helps determine the representativeness of the sample to the population within the sampling frame. The field interview teams collect the information to calculate response rates (i.e., numerator and denominator) on the tracking form. This underscores the need for the teams to be well-trained in using the tracking form and to be disciplined in tracking all households selected for interview, including those for which no interview was completed or no contact was made. To compile the tracking form data, tally the responses to each row on the tracking form. In some sections, it may be necessary to reconcile discrepancies (e.g., the interviewer may have appropriately selected both “no answer” and “door was answered” because the field interview team returned to complete a survey at a household previously selected but no one answered the door”). Select only the final designation (e.g., “door was answered”) for tally and disregard the prior visits. These tallies can then be entered into a spreadsheet, illustrated in Figure 10. Each column should represent a single cluster. If more than one tracking form was completed per cluster, consider forms for the same cluster cumulative and tally them onto the same column in the spreadsheet.

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO Figure 10. Sample dataset showing attempted and completed interviews per cluster

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO Calculation Response Rates The three common response rates included in the analysis are the completion rate, the cooperation rate, and the contact rate. The definition, calculation, challenges, and key considerations of these response rates are described in more detail in Table 11. Table 11. Common response rates Rate Definition Calculation Challenges Key Points Completion Represents how close interview teams came to collecting the goal number of interviews (typically n=210). Number of completed interviews / Goal number of interviews (typically n=210) Rates below 80% (typically n=168) result in an unacceptably low number to represent the sampling frame You should allot enough time for teams to complete a minimum of 80%. We recommend having a few teams available to return to low- responding clusters during follow-up day(s) Cooperation The proportion of households at which contact was made and agreed to complete an interview. Number of completed interviews / All households where contact was made* The higher the number of contacts made, the more the sample becomes one of convenience. It represents both the eligibility and the willingness of the community to complete the interview. Contact The proportion of households at which contact was attempted and agreed to complete an interview Number of completed interviews / Number of households where contact was attempted** The lower the contact rate, the more the sample becomes one of convenience at the second stage Higher contact rates indicate better representativeness of the sample to the population. Lower rates indicate that teams had to attempt interviews at many households to obtain the necessary interviews * including completed interviews, incomplete interviews, refusals, door was answered **including completed interviews, incomplete interviews, refusals, and appears as though someone is home but no answer, appears vacant, and nobody home after 1st , 2nd , or 3rd visit

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO R EPORTING THE R ESULTS Writing the report is the RNA’s final phase. Although it is the last phase, its planning and development should begin long before any data are collected in the field. An essential element to this process is to plan for the report in advance. During the preparation phase of the RNA , you should identify several foundational aspects of the report, such as who will write the report, who is the target audience, and what analysis you will include. Exploring these aspects with the RNA assessment team at the onset will allow the team to turnaround the reports quickly to key stakeholders (e.g., within 48 hours). We recommended the following two reports: • Preliminary Report or presentation provided to key stakeholders within a day or two of data collection. This will allow partners to make quick, better-informed decisions and address any immediate needs • A final report that builds off of the preliminary report and includes additional data analysis and results. This report is distributed to a wider audience and is often provided within a few weeks of the RNA As with any report, results should be in a simple, easy-to-read format that reflects the original objectives of the RNA. Table 12 provides a series of key considerations when planning and creating reports. Table 12. Report considerations 29 Action Description Designate a point person Designate a person who will be in charge of the reports. This person should be familiar with the entire process and involved in the preliminary meetings, sampling, and analysis. Create table shells Begin the writing process before completion of data collection. Gather background information for the report and create table shells that will expedite the development of the report. Identify audience needs and tailor content When drafting the report, consider the target audience. Will you send the report to emergency managers, to epidemiologists, to politicians, to the media, or to all of these recipients? Consider how each audience digests information and tailor accordingly. Use the audience to determine the format, how much 29 ibid

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO information to include, the report length, and the technical detail. For example: • Emergency managers may find the projected number of HH useful (e.g., 5k households need tarps, 4k households need medication, etc.). This information is probably best displayed in a bulleted list. • Epidemiologists will likely want the most technical information, including full tables with all the data (including all results, missing data, confidence Intervals, etc.). Politicians/media may need a written paragraph of overall results or bulleted format of key points. Graphs/charts might be more effective rather than lengthy tables. Recommend actions based on results Present results in an action-prompting format. Actions derive from the current health status or needs of the population as presented in the results. Actions can include, but are not limited to, prioritization of resources, public health messaging, enhanced health surveillance, or public health interventions. Preliminary Field Report The preliminary field report is a draft version of the final report. The preliminary report contains the initial results and recommendations to share with stakeholders who need time sensitive information. We strongly recommend that you inform partners and stakeholders of the preliminary results as soon as possible to ensure timely responses by other agencies and maximum transparency to the public. Given that the report has not been finalized, that fact should clearly appear on each page and should be clearly communicated to show that the report is preliminary in nature. Therefore, limited distribution of the preliminary report may be in order. Table 13 provides a common structure for the preliminary and final report. Table 13. Common reporting structure 30 Section Description Introduction Describe the purpose of the report, including the background of the disaster (e.g., date occurred, affected population) and details of the RNA request (e.g., who requested, who conducted, timing of assessment) and the RNA’s aims and objectives. Approach or methodology Provide the audience with an overview of the methodology used to produce those results that appear in the results section. Include 30 ibid

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO a simple description of the sampling frame (i.e., target population), the two-stage cluster design, the questionnaire, the number and training of field interview teams, and the data analysis procedures. Results Include the three response rates discussed earlier in this lesson, the number of households represented by the sampling frame, and the RNA’s main findings. Remember the audience needs and expectations and tailor the level of detail and content accordingly. Conclusions Recap the main findings, the RNA’s limitations, and the preliminary recommendations as based on the initial conclusions. Recommendations should be specific, objective, targeted, and tangible, thus enabling them to prompt action. Attachments If available, include other report types or supporting documentation in conjunction with the preliminary report, such as bulleted lists of highlights or press releases. We recommend that a presentation accompany the delivery of the preliminary report. Invite to the meeting all partners and appropriate community organizations. The meeting should mirror the structure of the preliminary report and include background, method, results, and conclusion sections. Give consideration to the best way to present the results in an oral presentation. An oral presentation is also a good opportunity to discuss implementation of the recommendations, next steps, and lessons learned, and to solicit input from a broad group of stakeholders Final Report The final report includes additional, detailed information regarding the RNA results. Release of the final report and its distribution to a wider audience is customary within several weeks after release of the preliminary report. A final report will include the following new and updated sections from the preliminary report: • Executive summary – develop an executive summary that highlights the report outcomes in a short and easily digestible format that also contains the report’s key results and conclusions • Updated results section – include key findings from the preliminary report as well as any additional analyses not included in the preliminary report (e.g., qualitative data or analysis of open-ended questions, if applicable). The results may include weighted and nonweighted frequencies, weight and nonweighted percentages, and confidence intervals that correct any inaccuracies from the preliminary report. • Updated conclusion section – include discussion from the preliminary report and incorporate any feedback received, additional information, and final

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO recommendations created in collaboration with partners. Also include any actions taken as a result of the RNA (e.g., enhanced public health messaging) • Acknowledgements section – include in the final report all individuals and organizations that helped with each stage of the RNA After the report is finalized, provide a final copy to all partners, including state and local public health authorities, emergency operations centers, incident command staff, and other agencies

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO Appendix A. Flood assessment form, June 2010 Question Response Q1. Date (MM/DD/YY) Q2. Interviewer Name Q3. Team Number: Q4 . County: Q5. Cluster number: Q6. How many people lived in your household before the flood? Number: _____ Q7. How many people slept here last night? Number: _____ Q8. How many in your household were children less than 2 years old? Number: _____ Q9 . How many in your household were 65 years of age or older? Number: _____ Q10. If one or more children are currently living in the household, do you have access to enough diapers and formula for 7 days (if needed)? □Yes □No □Don’t Know □Ref □NA Q11 . Is anyone in your household pregnant? Q11a. If YES , how many? □Yes □No □Don’t Know □Ref □NA Number: _____ Q12 . Did your whole household evacuate your home to sleep somewhere else? Q12a. If YES , what day did your household evacuate your home? (MM/DD/YY) Q12b. If YES , what day did your household return (spend the first night back)? (MM/DD/YY) □Yes □No □Don’t Know □Ref Date evacuated: ___/___/___ Date returned: ___/___/___ Q13. What is the primary type of construction of this dwelling (majority of exterior as observed by interviewer)? 1=Brick/stone/concrete 2=Siding (e.g., wood, aluminum, vinyl) 88=Other, specify __________ Q14. Would you describe the damage to your home as _______________ Q14b. If DAMAGED , do you need a tarp? 1=None or minimal 2=Damaged and habitable 3=Damaged and uninhabitable, but repairable 4=NOT repairable □Yes □No □Don’t Know □Ref

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO Q15. Do you feel the home is safe to live in? Q15a . If NO , when do you expect this home to be safe to live in? □Yes □No □Don’t Know □Ref 1=Less than a month 2=1–3 months 3=4–6 months 4=More than 6 months 5=Never 99=DK Q16. Has your household noticed any increase in problems with the following? □Mosquitoes □Flies □Rats □No problems Q17. Does your household currently have running water? □Yes □No □Don’t Know □Ref Q18. Does your household have a working indoor toilet? Q18a . If NO , does your household have access to a functioning toilet? □Yes □No □Don’t Know □Ref □Yes □No □Don’t Know □Ref Q19. Are there screens on the windows in this residence? □Yes □No □Don’t Know □Ref □Broken screens Q20 . Does your household have electricity from the utility company? □Yes □No □Don’t Know □Ref Q21 . Has your household used a generator at any time since the flood? Q21a. If YES , where is or was the generator located? □Yes □No □Don’t Know □Ref 1=Inside 2=Garage/shed 3=Carport 4=Outside, greater than 25 feet from home 5=Outside, less than 25 feet from home 88=Other 99=DK Q22. Since the flood, has your household used any of the following appliances indoors? □Charcoal, wood or propane cooking stove □Other gasoline or diesel powered equipment specify : __________________ □No □Don’t Know □Ref Q23 . Where is your household getting your drinking water from? 1=Well 2=Public utility

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO 3=Bottled, supplied by Red Cross or other relief agency 4=Bottled, not supplied by relief agency 5=No drinking water 99=DK Q24 . Is your household treating your drinking water? Q24a. If YES , how is your household treating your drinking water? □Yes □No □Don’t Know □Ref 1=Boiling 2=Bottled 3=Chemical treatment 4=No treatment 88=Other Q25. Do you or your household members need clothes? □Yes □No □Don’t Know □Ref Q26. Has anyone living in this household ever been told by a doctor, nurse, or other healthcare professional that he or she has the following health problems? a. Asthma b. Emphysema c. Hypertension d. Heart disease e. Diabetes f. Physical disability g. Hearing impairment h. Visual impairment/legally blind i. Gastrointestinal illness j. Renal disease k. Neurological (stroke, seizures, Transient Ischemic Attack (TIA)) l. Dementia/Alzheimer’s m. Arthritis and joint pain n. Cancer o. Anemia p. Psychiatric disorders q. Other __________________________ Q27. Has any household member become ill since the flood? Q27a . If YES , how many household members have had the following since the flood: □Yes □No □Don’t Know □Ref ___a. Stomachache/diarrhea ___b. Dehydration/ heat stress

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO ___c. Respiratory (cough, flu) ___d. Hearing or visual impairment ___e. Injury/ Physical disability ___f. Stroke (heat, hypertension, other) ___g. Heart attack (chest pain) ___h. Chronic illness that worsened ___i. Skin condition/rash/sunburn ___j. Stress/sleep disturbance ___k. Inability to perform daily task ___l. Other ________________________ Q28. Has any household member been injured since the flood? Q28a. If YES , how many household members have been injured or died from the following since the flood? □Yes □No □Don’t Know □Ref ___a. Drowned ___b. Motor vehicle-related injury or other blunt trauma ___c. Electrical injury ___d. Burn, including sunburn ___e. CO poisoning ___f. Laceration, abrasion ___g. Bruise/contusion ___h. Impalement/puncture wound ___i. Brain injury/concussion ___j. Strain/sprain/dislocation ___k. Fracture ___l. Dog bite ___m. Insect bite ___n. chemical exposure ___o. Other__________________ Q29. Since the flood, has anyone in your household required medical care? □Yes □No □Don’t Know □Ref Q30 . Does anyone in your household need medical care or supplies now? If yes, collect referral information. □Yes □No □Don’t Know □Ref Q31 . Will anyone in your household need medical care or supplies in the next 3 days to 1 week? □Yes □No □Don’t Know □Ref Q32. Can everyone in your household get the medical care and supplies he or she needs? □Yes □No □Don’t Know □Ref 1=Lack of transportation

D ISASTER P REPAREDNESS & R ESPONSE T RAINING : M ODULE T WO Q32a. If NO , what is preventing you/them from getting the medical attention you/they need? 2=No medical services available 3=Financial reasons 4=Debris or damage roads 5=Afraid to travel 6=Other:____________ Q33. Has any household member died during or after the flood? □Yes □No □Don’t Know □Ref Q34. Has anyone in your household had any of the following since the flood? (select ALL that apply) : 1=No symptoms 2=Difficulty concentrating 3=Trouble sleeping 4=Loss of appetite 5=Racing or pounding heartbeat 6=Agitated or frantic behavior 7=Violent behavior or threatening violence 8=Suicidal thoughts or attempts 9=Drug or alcohol intoxication or withdrawal 10= Bereavement: loss of family members, friends, neighbors 88=Others 99=Don’t Know Q35. Are emotional concerns preventing you or any household members from taking care of yourself/himself or others? □Yes □No □Don’t Know □Ref Q36. What is your households greatest need right now? (select one) 1=No current needs 2=Food 3=Electricity 4=Water 5=Medical care 6=Medications 7=Transportation 8=Physical help with cleanup and repairs 9=Financial help with cleanup and repairs 10=Trash removal 88=Other, specify _____________ 99=Don’t Know

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

CDC Module 2 Lesson 4

Related Documents