April 2017 Question 2
2. The study noted that the low SES population that typically uses food pantries is very difficult to reach, and therefore is likely underrepresented in national surveys and SNAP participants. How might this impact our understanding on diet quality among low SES groups? How would you propose researchers try to reach these groups?
posted by Unknown @ 7:46 AM
23 comments
23 Comments:
If the low SES population is difficult to reach, and therefore underrepresented in national surveys, the data we have on the low SES population may be different than the reality. As the data we have may be incorrect, we may not truly know whether they are nutrient-deficient, or which part of their diet actually needs to change. This could impact our focus on low SES populations.
For example, perhaps the current data says that low SES populations are deficient in calcium, but in reality, the majority consume an adequate amount. As the data is not representative of the entire low SES population, researchers were unaware of the truth. If this were the case, RDs may be pushing calcium on the low-SES population when they should really be focusing on another nutrient.
To reach this segment of the population, I think it’s important to go to them, wherever they are. If they are using food pantries, perhaps it would be beneficial to ask the food pantry users if they are willing to fill out a survey while they wait for their bag of groceries.
Most of this question stems back to a question of how can we get in contact with this group of individuals specifically. When thinking of different places a research might be able to contact a low SES family or individual, I wonder if there could be a survey or questionnaire type document added into the free and reduced lunch forms. This way the family could answer for all members of their household and more data might be collected at once. Families are also likely to fill out all papers of those packets in efforts to get their child access to food, even if the particular survey was only for research and not required.
Another location that may be able to screen or ask nutrition questions of a low income population may be at community clinics or free health centers. Asking questions on the documents a patient fills out when checking in may allow for that, or having the physician ask the patient during their exam may also be useful to acquire nutrition intake information.
Overall, I think finding a targeted area that will capture that specific population is the key to getting better response numbers.
It seems that we do not have a clear picture as to the diet quality among low SES groups. Besides the fact that this population is difficult to reach, I believe another factor is that there is no way to pinpoint intake if the foods available to them are so inconsistent (ex: eggplants donated from Kroger, French bread donated from local bakery, random donation from X, etc). This means that researchers would need to create a large, longitudinal study that encompasses different ethnicities, ages, neighborhoods/counties/states. Now how to reach these people? Similar to Lauren’s idea of conversations with the doctors/staff at health clinics, I thought that perhaps qualitative data could be collected at food pantry locations- via conversations while passing out the food or while the individuals are waiting in line.
With the low SES population likely being underrepresented in national data, I don't believe it would skew our current data drastically. With that said, it is difficult to really know what foods people are choosing, or buying. Food stamps aren't as strict with what products people can buy compared to WIC. Low SES families may be purchasing wholesome food for their families, while others may be choosing the calorie-dense, low-nutrient food items. When determining how to reach these populations, what comes to my mind is to be present at the main hubs in the community they are found such as food banks, soup kitchens, federally qualified healthcare centers and community centers. Gathering data on this population needs to be convenient for them, and going to them will ensure you can get the information needed.
I think something important to point out that was brought up in the limitations section is this systematic review did not include any studies that represent food pantries on a national level, which may have skewed the results. I believe more studies need to be completed assessing diet quality of food pantry users on a national level so we can obtain data that is more easily generalized. Additionally, I think longitudinal studies would provide more useful and reliable information. However, as the study noted this population is difficult to reach, making longitudinal data difficult to collect. I believe when it comes to research, in person recruitment/data collection and incentives are the best methods for data collection, although this often isn't far reaching and requires more time. Some individuals in this population may not have access to online resources, so email or online survey may not be the best option.
Lauren, I think your idea of recruiting at community clinics and health centers is a great suggestion. Patients may have time while waiting to be called to see the physician and would be a great opportunity for researchers to approach potential participants.
Nikki and Kaitlyn, I think recruiting at the food pantry itself is a great a idea, too! It’s obvious that if they are at the food bank to receive donations they currently participate in the food pantry and would most likely qualify for the study. It would also be a good idea to include some type of incentive if the study was longitudinal.
Lauren, I like your ideas of adding a questionnaire to the free and reduced lunch forms for school aged children and at free health clinics. I think these are great ways to reach this segment of the population!
I also agree with Tessa that in-person tactics will probably be the most effective as this population may not have access to email or online surveys. In-person contact will take more resources, but I believe understanding this population is vital to helping them and therefore worth the investment of extra resources.
I would have to agree with Nikki, that despite perhaps not having the most representative sample of the SES population, I don’t think it would drastically change the outcome of the results, or maybe more importantly, our focus and work as R.D.s with this population. It makes sense that low SES individuals most likely do not have ample resources to provide for all of their family’s needs and thus nutritional intake quality may suffer.
In terms of reaching out to SES groups and individuals and including them in our research, I would agree with Erin and Lauren that we have to go to “where they are”. Food pantries are obviously a good place to start, but so are other community- or need-based programs like Lauren mentioned. Similar to her idea of using free and reduced-lunch applications to get in touch with this population, visiting early-childhood or other government-funded “Head Start” programs for low-income families would be another viable option. At the food pantry I was involved with in the past, we actually held a picnic lunch at Head Start for the children and their parents and used it as an opportunity to provide them with a free lunch, but also a nutrition education activity and to gather some data for county-based government research on this low SES population.
Boys and Girls clubs may be another place to visit to get in touch with this population and in turn put forth some nutrition education efforts, as well.
I think having a low SES population is a limitation of this study. However, I agree with Nikki that I do not believe that the results would be skewed too drastically. When suggesting how to reach this population, I would agree with many of the other Interns that in person contact at the location they are receiving their food from would be the best way to collect data. I think that at food banks they could also require that they complete a small survey before collecting their food.
Tessa, I really like your idea of some type of incentive to participate in data collection. I think that offering incentive can increase participation in many studies and would be a good idea to try.
The noted difficulty in reaching the low SES population that uses food pantries like leads to underrepresentation in surveys and research studies. The actual severity and prevalence of poor diet quality in food pantry users is likely worse than reported due to this probable underrepresentation. With the actual problem possibly being worse than reported, the urgency of a larger solution at the national level should be of critical concern. To reach these difficult-to-reach and likely underrepresented groups, they must be intercepted at the point at which they receive community resources. As others have mentioned, making contact with these people when they come to the food pantry, community meal site, WIC office, health clinics, etc is crucial. Once they people leave the resource site, the chances are slim that a researcher/surveyor will be able to contact them and elicit information. Additional challenges this population may face include transportation difficulties, working long hours and/or multiple jobs, challenges finding childcare, limited healthcare access, and possibly homelessness. These examples of difficult life circumstances can further complicate communication with the low SES population discusses.
The fact that the low SES population is very difficult to reach and likely underrepresented in national surveys and SNAP participants may have an impact on our understanding on diet quality because the research is not reflective of the trends of the population as a whole. In fact, I would expect the actual data would reflect a dietary quality that is much lower than what we would expect. I have been instructed the best way to reach a "difficult to reach population" is to go into the places they already reside. One would need to go to food pantries, donation centers, libraries, etc. or where ever it is the population being studied already travels to. It is difficult for them to come to you so you must go to them. In the end, I believe researchers would have more participants in their study that would benefit their research and increase accuracy.
With this population being difficult to reach, there is a good chance that we do not have accurate data to represent the entire population. Perhaps those that are more likely to participate in studies also have a better or worse diet than the general population. Hopefully the ones that are sampled do follow the patterns of the entire population, but it is very hard to determine that. I think Kaitlin had an excellent point about the inconsistencies of intakes with this population as much of their food supply changes based on donations. Like I said in my previous post, I had a rotation at a food pantry, and I would help sort donations from places like War-Mart. Their donations were always different. One day it may be a large assortment of cakes and pastries from the bakery, and another day they will send a large shipment of whole grain breads. The nutrient contents are vastly different depending on what they have to offer that day.
In order to reach more participants, I agree with everyone that said we should go to where the people are. Reaching out to food pantries would be a great place to start. Lauren also mentioned community clinics. Places such as WIC would be an option. From my little experience interviewing at a WIC location, I believe the patients make their way through different stations of their visit, talking to the nurse about health, talking to the dietitian about nutrition, etc. through the end of their appointment where they are sent with food stamps and other types of assistance programs. This could be the time to provide the survey and request that they fill it out before the end of their appointment. Incentives would also be really beneficial to promote more responses.
As stated above, the difficulty in reaching the low SES population may skew the diet quality data. I would expect the actual diet quality to be worse than reported; however, I also feel the data should be interpreted with important context in mind. In particular, it should be kept in mind that only six of the studies used were conducted in the U.S., and the level of support the U.S. government gives to low SES families may differ from the other countries included in the study.
Regardless of international differences, I would suspect that low SES individuals would have several similarities (at least in "Western" countries)--namely a lack of transportation and lower purchasing power. Thus, I think the "go to them" strategy is likely valid for more accurate data collection. I would like to see a similar study featuring only the American samples, as I suspect they would have lower diet quality than say low SES French or German families simply for cultural reasons, yet alone food supply infrastructure.
The data we have on diet quality is not very useful if it does not provide a good representation of the population in interest.
I'm on board with Michael and the others. I also believe that you would need to go to that population in order to reach them for accurate, valid information. Researchers could go to food pantries with surveys, questionnaires, and/or diet recalls and ask for users to participate. An incentive is always helpful in order to thank the participant and increase participation, so perhaps they could offer a food item, a coupon, or even a small amount of money.
I really like Lauren's unique ideas of surveying this population. Adding questionnaires to forms such as Free and Reduced Lunch forms would most likely increase participation due to the reasons she stated.
If we are unable to reach the low SES population for surveys, then the results from these surveys are unreliable. If we only use these surveys to form our education intervention then we may be targeting the wrong thing or we may be focusing on one problem where there are several others we should also consider. It is important to put effort into getting more representative data on this population if we want to assist them to the best of our abilities.
If they are difficult to reach then the solution would be to go to them. They obviously go to food pantries so that is where surveys should be given at. Low response rates may be due to the fact that English is not their first language or they have poor literacy skills. A confidential meeting area to verbally ask questions to participants may help improve the response rate and give a better representation of what is actually occurring in this population.
This greatly impacts the results and understanding of different SES in relation to diet composition. How can we truly understand a population if we aren't getting an accurate representation of them. Like I said in my post for question one, I think the most effective way of gathering information for low SES who use food banks/pantries would be to physically go to these locations during operational hours and talk to the individuals who attend. This would take a lot of time and effort, but it would help to better represent this population and eliminate the excuse that they are difficult to reach. I personally do not think this population is difficult to reach, but it is more of a concern for the researchers pushing their comfort zones.
Everyone has mentioned numerous ideas for ways to collect data from the SES population. I agree that surveys and questionnaires should be easily accessible for higher response rates. As Tessa mentioned, it is difficult to know whether or not people will have the necessary resources to complete an online survey, so I think the idea of gathering information at local food pantries or free health clinics is most ideal. In order for this group to not be underrepresented, collecting data should be a top priority. My only question is that if researchers knew that the SES population who use food pantries is underrepresented, then why have they gone this long without trying to collect data at places they know this group visits? To me, it is almost common sense to think about collecting data about diet quality of the SES population directly at a food pantry. I like Jamie's idea of providing an incentive to increase response rates and participation.
As the study noted it is difficult to reach the low SES population that utilizes food pantries, leading to underrepresentation of this population in research. Since these populations are often underrepresented in research it is likely that the severity of poor diets in the SES population is worse that the suggests. Often individuals of low SES do not have the internet or a home phone in their residences. Therefore, the best way to reach these groups would be through in person contact. This contact could include providing education through schools or physicians/clinics or direct contact at with individuals at food banks or door-to-door.
I find the proposed limitation to be significant. The lack of direct reliable sampling of the SES population results in the development of intervention programs based only on the conjecture of public health experts. Despite the best intentions it is possible efforts and resources are being directed to the wrong causes.
As identified from the flawed collection methods of the 15 studies of this review nutritionally important absences exist among food pantry users, which may be present to a greater extent among the SES population.
In conducting research on this topic I would attempt to a convenience sample visiting areas of a central area of the desired SES population and offer a participation incentive. Areas to target may be food pantries, public centers, libraries, convenience stores.
Included in the questions should be a disqualifier of the estimated average income to sort out surveys which were completed by members of the public which do not fit into the SES population but were otherwise captured in the convenience sample.
Further a quick nutrition assessment tool would be needed such as food frequency survey.
An underrepresented population is going to have a large affect on the validity of data found within that population. As Erin stated, because diet quality encompasses many different aspects of the diet, our interpretation of diet quality in a low SES population may be headed in the wrong direction which may cause us to overlook the real problem if we are not careful. I think that data from studies like this give us a good idea of what to look for but we should always do our own research on the surrounding population that we are serving.
I believe all of you share great ideas for contacting individuals. The idea of offering an incentive could go a long way. Also, Kandice makes a good observation regarding lack of internet and phone connection. We would have to be proactive and go to schools, clinics, food pantries, after school programs, etc. in order to obtain accurate information.
I think that this is an interesting and very accurate concern regarding the representation of specific groups that utilize food pantries. During my rotation at the Diabetes Care Center in Peoria (UPH), Holly and I had an in-depth discussion regarding the difficulty reaching specific populations for diabetes education. We talked with an individual who ran a large food pantry in the area regarding how he effectively disseminates information to participants. We learned that it may be useful to reach these populations based upon cultural norms; in the instance of these particular populations, a potential channel of communication was through churches. I feel that discussions like these may help to better reach populations for the national surveys. At this time, I do feel that limited outreach to lower SES populations for representation in surveys does skew data. As mentioned in the article, there is a paradox between obesity and low SES status that is not fully understood. I think that if we can reach more of this population, we can get a better grasp on the full nutritional quality of the American diet.
I think Kandice brings up a valid point. It is important to acknowledge barriers faced by lower SES populations, especially regarding communication. Researchers will not be able to effectively reach individuals if they do not have a phone or internet access. Understanding the populations targeted will help to better access accurate data.
Tessa, I think you made a good point about using incentives to help increase participation rates of low-SES individuals. However, as a researcher one needs to make sure that the incentives are not large enough that ethics or bias/error questions don’t become raised.
Also, I think Alyssa brought up a good point about the inconsistency of products donated to food banks. Many individuals of lower-SES rely of food banks for assistance, making it challenging for participants to ensure they have a balance diet. For example, if on the day an individual is at the pantry to pick of their weekly bag the food, the bank may only have an assortment of cakes and pastries and no bread because that is all that was donated that week.
Many of you had great suggestions. I think contacting the food pantries and the health clinics would be a great place to start. Another thing to consider is the willingness and time commitment these individuals would be likely to make in order to answer a survey. Traditionally, FFQ tend to be long and may be considered tedious. This may deter many individuals from participating altogether. It may not be that we aren't reaching these individuals, but rather the tools used to assess their data is the problem.
Lauren your point of introducing surveys where consumers already expect to be releasing information is a valuable suggestion. Other ideas may be public institution registrations, meals on wheels service enrollment or utility aid grants.
Post a Comment
<< Home