Responses to OMB questions concerning NIOSH survey

Responses to OMB questions concerning the NIOSH survey

(0535-0235).

1) The justification for focusing on minority operated farms is not clear.  Neither the UDSA Dec 2002 report (2000) nor the NIOSH 2007 flyer (2003 data) nor the supporting statement compare childhood injury rates from minority and non-minority operated farms.  Is there a major difference?  Given that NAAS/NIOSH has collected such minority data in the past, have you seen trends in the increase or decrease in accidents at minority operated farms that are different from those of non-minority operated farms?  

 

The justification for focusing on minority operated farms is that they are such a small portion of the overall farming population (less than 5%) that it is not possible to track youth or adult injury and asthma problems for these groups from a random sample of 50,000 farming operations.  In addition to the collection of injury information, NIOSH needs and uses the demographic information on farm youth living or working on these minority farms to determine the number of minority youth at risk for agricultural injuries and asthma. These demographic data are also used to assess the number of minority youth operating farm tractors, operating All-Terrain Vehicles, and riding horses on their family farms—activities that are highly associated with injury to youth on farms. Again, the collection of these data is not possible without targeting these groups with a specific minority-focused survey.  With respect to the question on whether minority farm operations are different from the general farming population, data from previous surveys indicate that differences do exist.  Specific examples include: that Native American youth have significantly higher farm injury rates than seen for youth in the general farming population; that Hispanic youth have higher work-related farm injury rates than seen for youth in the general farming population; that all minority youth living on farms have significantly higher prevalence rates of asthma compared to the general US population.  In addition to documenting these differences, NIOSH is using these surveys as part of an overall public health surveillance program to address youth farm injuries, asthma to youth on farms, and adult work-related farm injuries.  Because of this, it is critical to continue monitoring these minority farm operations to allow NIOSH to track changes over time for these three end points.

 

2) The justification statement focuses on minority operated farms, but the rest of the supporting statement does not specifically refer to minorities.  The only other specific mention we find is the sentence in item 12 of Part A, which says "In order to receive enough data to be able to publish detailed results by racial minority groups and Hispanic, it was determined that the samples size needed to remain the same (as last time).   Furthermore, we did not find a 'screener' for identifying minorities and Part B does not discuss minorities. This suggests that you are seeking clearance for a study to be administered to a random sample of the entire farm sector, from which you hope to be able to also say something about minorities.  This does not seem consistent with the justification, but if that is the case, please provide the analysis that supports your estimate that a sample of 50,000 is sufficient to give you a suitably diverse and representative sample of minority farms.  If this is not the strategy that is being proposed, please provide more information.  

 

The package as submitted is intended to obtain a three-year clearance that would cover all NIOSH childhood agricultural injury surveys conducted by NASS.  This includes the minority-specific and the general farming population surveys.  For 2009, the survey will sample only minority farm operators identified in the 2007 Census of Agriculture.  No non-minority farm operators will be included in the sample.  In 2010, the sample of 50,000 farms will cover all farm operators in the NASS sampling frame. 

Minority farms are selected from the NASS list frame based on indicators designating the farm as operated by a minority.

 

3)  A screener is referred to in Section A.12 within the context of screening out 'no children' - we did not see this in the submission screener either.  

 

The questionnaire has three questions to the operator.  Do they have youth living in the household?  Do they hire any youth?  Did any youth visit the farm?  If the operator says "no" to these three questions, then they will exit the youth injury portion of the questionnaire and not be asked any questions about youth asthma or youth injuries.  They potentially would still answer the adult portion of the questionnaire if they are selected for that survey as well.

 

4)  A pilot study is referred to (sample of 500).  What is the purpose of the pilot? (i.e., what are the goals, what sorts of changes to you plan as a result, etc).  

 

The term “Pilot study” is used loosely to describe what is a pre-test of the Blaise data collection instrument. This is done prior to data collection to ensure nothing has been overlooked developing the software, and that data collection procedures are sound.

 

5) Is the same survey that was administered in prior years?  Please enumerate the list of changes since the last approval.  

 

The survey is essentially the same as previous years.  Changes include dropping certain farm demographic questions such as "acreage," "type of farm" and "value of sales" because NASS will provide these variables to NIOSH from their sample frame database. This was done to remove unnecessary questions from the survey.  For all surveys, the following questions have been added: 

• Is there a completely enclosed, fenced off play area on your farm for children?                                                                                                                         

• Do you have access to licensed, off-farm child care?  (if yes)  How often do you utilize this service?

For the adult survey, the following asthma questions were added:

• Have/Has (you/the farm operator) ever been told by a doctor, nurse, or other health professional that (you/they) had asthma?                                               

• How old (were you/was the farm operator) when asthma was diagnosed?                                                                                                                      

• (Do you/Does the farm operator) still have asthma?  

                                                                                                                    

• (Have you/was the farm operator) ever told by a doctor, nurse, or other health professional that (your/their) asthma was related to (your/their) work on the farm?                                                                                                                                                                      

• Did (you/the farm operator) have one or more asthma attacks requiring the use of an inhaler or other medical treatment in the last 12 months?    

• Did any such asthma attack occur while doing farm work?

• Did (you/the farm operator) have a serious asthma attack that required an emergency room visit, hospitalization, or other professional medical attention in the last 12 months?                                                                                                                                     

• Did any such asthma attack occur while doing farm work? 

6) The burden table is based on a sample of 50,000 (half for children to adults); Part B talks predicts a 60% complete rate (using the OMB formula).  If this is the case, why do you estimate full burden for 40,000 completed surveys?  Why would you calculate burden for out of business farms?  

While the OMB equation for response rate comes to 60%, the actual number of telephone contacts made will be near 40,000 individuals.  Out of business farms are included in the estimate of burden because NASS does not know they are out of business until they are called.  In addition, even if a respondent says they are out of business, NASS asks a series of 5 questions:

- Did you grow any crops or cut hay in 2008?

- Is any of the land in this operation cropland? (Including idle cropland and cropland in government programs such as CRP, etc.)

- In 2008, did you have any whole grains, oilseeds, or hay stored on this operation? 

- Do you have facilities for storing whole grains or oilseeds?

- Do you own or raise any livestock or poultry?   

 If the person answers "yes" to any of these, NASS considers them still in business and they would continue with the interview.  Even if they answer "no" to any of these questions, the respondent has incurred a time burden. 

7) With respect to the response rate, what sort of non-response bias do you have planned?  How will this get at whether the minority population is more or less representative than the non-minority population?  

The current plan to assess the impact of non-response is to look at the demographic variables for these non-respondents (value of sale, acreage, type of farm, race of the operator, Hispanic origin of the operator, etc...), and see if there are any significant differences between the non-respondents compared to the 2007 Census of Agriculture demographics for these minority operators.  In addition, the survey respondents will be benchmarked to the 2007 Census of Agricultural by minority category (race or Hispanic origin) and region of the US.  Additional benchmarking to the 2007 Census of Agriculture (e.g.., adding value of sales to the benchmarking process) may be considered if the non-respondents do differ significantly from the respondents. 

 

8) With respect to the burden time per farm - 5 min (child only) or 10 min (both) - that seems ambitious for a 22 page survey.  Is this based on an estimate that X percent will screen out before the first minute?  If so, what is the X?  

For the youth survey, we project only 30% of the farms will have household youth or hired youth on their farm, so only 9000 farms provide any youth demographic details.  We also project that only 1% of the farms will have a childhood farm injury, which means only 300 farms are expected to answer the childhood injury questions.  For adult injuries, we expect only 4% of the farms to report injuries, so that only 600 farms will have any additional burden from the adult injury questions.