Challenges In Conflict Data Collection: Assessing The Spatial Characteristics Of Nutritional Status In Erbil, Kurdistan, Iraq
BY HAMDIA AHMAD, TABAN K. RASHEED, LISA R. PAWLOSKI, AND KEVIN M. CURTIN
It has been well-documented that significant difficulties arise when conducting field research in conflict areas.1 This article documents the process of collecting nutritional, anthropometric, and spatial data in a conflict region. Specifically, data were collected from more than 1,000 participants in the city of Erbil, Kurdistan, Iraq in the summer of 2011.
Significant change has occurred concerning the political, economic, and social landscape within Iraq in the past thirty years. Kurdistan, while relatively isolated from the second Gulf War, has not been immune to the negative social and economic effects of continuing political upheaval. Throughout Iraq and Kurdistan, conflict has been a daily occurrence since the Iraq-Iran war from 1980 to 1988. Since then, Kurdistan’s healthcare system has suffered significant economic losses and neglect. While some external cooperation and internal development has allowed for economic improvement from the second Gulf War starting in 2001 and from the implementation of the “No fly zone” in 1991 to protect the Kurdish population from the Iraqi military, there are still great disparities in health status as well as healthcare among the people of Kurdistan. Many continue to have restricted access to health facilities or food. In terms of health status, while there is little detailed information available, experts suggest that Kurdistan has begun to face problems similar to those experienced in other transitional countries, where obesity and chronic illnesses are beginning to emerge, in addition to high rates of diseases more commonly found in low income countries including infectious disease.2
As Kurdistan continues to develop, there is now an even more urgent need to describe the health and nutritional situation among the Kurdish people, particularly to have a better understanding of the specific problems, disparities, and social and geographic determinants of health status. As different parts of the region are affected by varying factors (i.e. refugees, conflict, national borders), understanding geographic patterns is critical to understanding the determinants of health status as well as implications for interventions and policy. Such analyses are important to understanding the causes and emergence of obesity and chronic illnesses in similar transitional and conflict regions.
One common and effective means of determining the health situation among a population is to examine nutritional indicators, as nutritional status provides a basic understanding of the health of a population, such that populations experiencing undernutrition are typically at risk for infectious diseases, while indicators of overnutrition and obesity suggest risk for chronic illnesses. As women are typically at greater risk of malnutrition in transitional countries this study presents the process of conducting data collection for a pilot study concerning nutritional and geographic indicators among healthy adult women living in Erbil, Kurdistan.
In order to gain a greater understanding of nutritional status among women in a conflict area, the following four initial research questions were formulated:
1. Describe the nutritional status of the sample of healthy women from Erbil, Iraq.
2.What are the relationships between BMI and socioeconomic status, activity status, commute time, and diet among women in Erbil, Iraq?
3. Does a spatial point pattern analysis of overweight/obesity show variability among the population or significant clustering in specific locations?
4. What is the geographic distribution of overweight/obese women in Erbil, Iraq?
The researchers received funding from the Center for Global Studies in order to work with researchers in Kurdistan to plan and execute the data collection. The researchers arrived in Erbil in the spring of 2011 to 1) develop the surveys; 2) develop a plan for spatial data collection; 3) develop a time-line for data collection; 4) develop a recruitment plan for participants 5) obtain proper clearances; and 6) train researchers on nutrition assessment techniques. Data were collected in the summer of 2011.
Surveys were designed to collect data on family structure, general demographic information, socioeconomic status, dietary intake, and activity. Multiple meetings were held to develop culturally sensitive questions that would capture the most accurate information. The surveys were generated first in English then translated into Kurdish and back-translated into English to ensure validity. The U.S. team worked closely with the Kurdish team to understand the typical dietary intake and activity patterns of Kurdish women in order to develop the most appropriate questions. For example, milk is rarely consumed, and thus determined not to be a significant part of the diet to assess. The researchers visited Kurdish homes to better understand the family structures, diet, and daily activities of the populace.
Spatial data development
In typical spatial survey research exercises, participants are asked to self-identify their location. They are generally asked to provide a home location, but work locations or other locations of significant activity may be requested. These requests are generally made by the addition of questions to the survey instrument, and the response is expected to be in the form of a standardized street address, or a ZIP code, or a coordinate in some other well-defined spatial referencing system. These responses can then be entered and georeferenced relatively easily using off-the-shelf geographic information systems (GIS). Once entered into the GIS, subsequent spatial analysis can be readily performed.
However in less developed, or conflict areas, the spatial data infrastructure is frequently not as well-defined and participants cannot always provide their location in a standardized form. This was true of the participants in the Erbil study. While there are identifying numbers on homes in Erbil, the Kurdish research team noted that these numbers were not used for delivery purposes and there was no known GIS reference database to which these numbers could be related. Furthermore, many people would not have the number for their home memorized in order to add it to the survey questionnaire. Moreover, in Erbil, postal service is not provided in a regular, house-by-house, daily service, and therefore formal street addresses are not as frequently used.
Since determining the location of the survey participants was critical to answering many of the posited research questions, several potential alternative methods were considered. One possibility was to visit the homes of each participant and collect the location using a Global Positioning System (GPS) collection device. Given that it was intended that the researchers would collect data at a central location, this option would have multiplied the time, effort, and resources needed to collect the data. It also seemed imprudent given the nature of the ongoing conflict in the area. Another alternative was to ask participants to carry a GPS device with them to their homes, capture the location data and return the unit to the data collection center. Given that only a few GPS units were available, and that each participant would have to be trained in the use of the device, this option also was deemed to be unworkable. Finally, there was some ambiguity as to how GPS devices could be used, and there was some concern that this data collection process could be misinterpreted as a military exercise.
The third alternative—and the one that was ultimately chosen—was suggested by the Kurdish research team when they found a highly detailed street map of the city and environs. The map also had underlying imagery showing the buildings and other land uses. The map included a great deal of annotation that would allow users to find their own location relatively quickly (Figure 1). It was decided that each participant would be asked to locate their home on the map with as much precision as possible, and that a label would be placed on the map indicating the identification number of the participant. A second label with the same number would be placed on the survey.
Overview and inset of the location map.
Developing a timeline for data collection was a challenge as there were many time sensitive issues that could inhibit data collection. Primarily, the researchers were planning to collect data from women working and attending the university, so the university schedule greatly impacted when data could be collected, as many went on vacation after July. Further, the researchers hoped to collect most of the data before the temperatures climbed too high, again during the summer months. Additionally, the researchers had to allow for some flexibility in the timeline as many issues including, protests, elections, holidays, and severe weather events could potentially hinder the data collection process.
To better facilitate data collection, additional research assistants were employed to collect data and recruit participants. The participants were limited to those living in Erbil, as there was not sufficient time or funds to travel multiple times to rural regions in order to collect data.
Recruitment of Participants:
As the timing of data collection was critical we used convenience and snowball sampling methods. The researchers recruited faculty, staff, and students attending the University of Salahhadin and asked them to come to specific convenient sites to be measured and respond to questionnaires. Figure 2 shows one of the researchers collecting anthropometric data at the University. Ultimately, 1,023 participants were recruited, measured, and surveyed. Only healthy (non-pregnant) women who were faculty, staff, or students at the University of Salahhadin were asked to participate in the study.
Anthropometric Measurement Collection
Institutional Review Board:
Researchers provided participants with information about the significance of the study and details about the methods. Participants were given an informed consent form to sign stating that they were not required to participate and could drop out of the study at any time. While it is critical to have a Human Subjects Review Board review such a study to ensure ethical procedures, no such process existed at the University of Salahhadin. Thus the study was reviewed by the George Mason University Human Subjects Review Board (HSRB), with a letter of review and support provided by the administration of the University of Salahhadin.
Having a U.S. institution lead such a review created significant cultural barriers and challenges. For example, all researchers in Kurdistan were required to take a formal National Institutes of Health Collaborative Institutional Training Initiative (CITI) online training course. While this training is helpful to understanding ethical human subjects procedures, it is not culturally sensitive and not available in Kurdish. The Kurdish co-investigators felt uncomfortable with some of the requirements stated in the online training. One such example included the following statement: “For this Completion Report to be valid, the learner listed above must be affiliated with a CITI participating institution. Falsified information and unauthorized use of the CITI course site is unethical, and may be considered scientific misconduct by your institution).” This statement caused considerable angst and fear as the researcher was not officially affiliated faculty at George Mason University, yet was being required to take the course by the HSRB at George Mason.
A major effort of the project was to train the researchers in nutrition assessment and data collection. The training involved the measurement of nutritional indicators, including anthropometric measurements, blood pressure, and hemoglobin status. In order to respect cultural norms with regard to gender, training was conducted only by a female researcher. As the Kurdish team did not have the necessary equipment to conduct the assessments, the researchers purchased and donated much of the equipment as well as loaned out the more expensive equipment for the field collection exercise.
The loaned equipment was eventually shipped back to the United States. This effort took approximately one year to resolve as the shipping companies in Kurdistan were very hesitant to ship out the equipment to the United States. The researchers determined that labeling the equipment as “medical equipment” created concerns on the value and even safety regarding what was been shipped to the United States. Perhaps only in such conflict regions would a skinfold caliper arouse suspicions of terrorism-related activities.
The data collection process resulted in 1,023 subjects completing the survey, submitting to anthropometric measurement, and identifying their home locations. Details of the survey and spatial results are as follows:
Surveys were distributed to determine socioeconomic status, activity levels, and transportation means. For example, the household size was an average of 6.01 individuals and women had an average of 1.41 children.
Anthropometric indicators included height, weight, hip circumference, and waist circumference. Height was measured using a field portable anthropometer. Circumferences were measured using non-flexible measuring tape. Anthropometric measurements were collected based on methods described by Lohman, Roche, and Martorell (1988).3 The anthropometric data revealed indicators of both underweight and overweight among women. Mean values of waist/hip circumference and BMI were above recommended healthy indicators of 0.80 and 25 respectively. Indicators of underweight were found in 3.4% of the sample and 44% were found to be either overweight or obese.
Hemoglobin status, an indicator of anemia, was determined using the HemoCue b-Hemoglobin System with the assistance of a local nurse. This system assesses hemoglobin levels and requires a small blood sample from simple finger sticks.4 Hemoglobin values were higher than the WHO recommendation of 12.0 mg/dl.
Spatial Data Results
Participants referred to the large satellite image of Erbil City and were asked to place a marker where they lived (see Figure 1). Preprocessing of the Erbil location map included the scanning of the original double-sided map, appending the two sides into a single location map, enlarging the image as much as possible given the large-format printer limitations, and printing of 12 copies of the location map. Since it was expected that over 1,000 participants may be interviewed, multiple maps were necessary to allow all of the labels to be placed without excessive overlap.
When the maps were returned from Erbil, post-processing included scanning of the maps with the labels affixed, georeferencing the scanned images to known GIS layers of Erbil (specifically a street network database), and digitizing the locations indicated by the labels placed by the participants (see Figure 3).
Overview and inset of the location map with participant labels
DISCUSSION, CONCLUSIONS, AND FUTURE RESEARCH
There are very few thorough recent studies concerning the health and nutritional status among women in Kurdistan, and none which exploit the spatial dimension. Much of the health related literature in Iraq had been conducted in the southern regions prior to 1991, in which high rates of anemia and undernutrition were reported. These studies were conducted by UNICEF and only included children.5 Recently, there have been several studies reporting the nutritional status of women in Iraq but only a small number have been conducted in Kurdistan. In general, the recent studies in Iraq reveal that there is growing problem of overweight and obesity among Iraqi women in southern Iraq ((Mansour AA & Ajeel NA. 2009. “Parity is associated with increased waist circumference and other anthropometric indices of obesity”. Eating and Weight Disorders. 14: 50-55; Al-Tawil NG; Abdulla MM; Abdul Ameer AJ. 2007. “Prevalence of and factors associated with overweight and obesity among a group of Iraqi women”. Eastern Mediterranean Health Journal. 13: 420-429.)) Further, the Iraq Family Health Study (IFHS) conducted in 2007 revealed that 21.9% of women in Kurdistan have anemia compared with 38% in the rest of Iraq. While not broken down by region or sex, the IFHS also report 42% of the population of Kurdistan has hypertension and 14.3% have diabetes.6 While the IFHS is the most recent report with information concerning the health status of women in Kurdistan, the report is primarily focused on understanding reproductive health issues and there is little information concerning the specific situation of Kurdish women. Other studies aimed at examining the health status of women in Kurdistan primarily explore the issues of infectious diseases and Thalessemia (a form of anemia, closely related to sickle cell anemia). Thus, this study provides needed information concerning the nutritional status of women in Kurdistan.
This report outlines both the significant difficulties in performing data collection exercises in a specific conflict region: Kurdistan, Iraq. It is hoped that this report will inform future data collection efforts and allow future researchers to avoid similar pitfalls. More specifically, issues regarding spatial data collection and culturally sensitive fieldwork have been addressed. The data suggest that diet and household structure are determinants of obesity among women in Kurdistan, Iraq, but more work is to be done. Future research and analyses will address all of the research questions above, including using spatial statistics in a GIS environment to determine if there are significant clusters of nutritional characteristics.
This research was supported by the Center for Global Studies, and Provost’s Office, George Mason University, Fairfax, VA.
Hamdia Ahmad is a Lecturer in the Faculty of Nursing at Hawler Medical University College of Nursing.
Taban Rasheed is a Lecturer in the Department of Biology at Salahaddin University College of Science.
Lisa R. Pawloski is Professor and Chair of the Department of Nutrition and Food Science at George Mason University.
Kevin M. Curtin is an Associate Professor of Geography and GeoInformation Science at George Mason University.
LIST OF REFERENCES
Al-Tawil, N.G.; Abdulla MM; Abdul Ameer AJ. 2007. “Prevalence of and factors associated with overweight and obesity among a group of Iraqi women”. Eastern Mediterranean Health Journal. 13:420-429.
Bogin, B. 1999. Patterns of human growth. Cambridge: Cambridge University Press.
Burger, S.E., Pierre-Louis JN. 2002. How to assess anemia and use the Hemocue. New York, New York; Hellen Keller Worldwide.
Cohen, B.B. 2012. “Conducting evaluation in contested terrain: Challenges, methodology and approach in an American context”. Evaluation and Program Planning. 35:189-198.
Cohen, N, and Arieli, T. 2011. “Field research in conflict environments: Methodological challenges and snowball sampling”. Journal of Peace Research. 48:423-435.
Eveleth, P.B., and Tanner JM. 1990. World wide variation in human growth (2nd. ed.). Cambridge. Cambridge University Press.
Field, J.O. & Russell RM. 1992 “Nutrition mission to Iraq for UNICEF”. Nutrition Reviews. 50:41-46.
Hatløy, A., Torheim, L. E. & Oshaug, A. 1998. “Food variety—a good indicator of nutritional adequacy of the diet? A case study from an urban area in Mali, West Africa”. European Journal of Clinical Nutrition. 52, 891–898.
Lohman, G, Roche, AF, & Martorell, R. 1988. Anthropometric Standardization Reference Manual. Chicago, IL; Human Kinetics Books.
Mansour, A.A. & Ajeel NA. 2009. “Parity is associated with increased waist circumference and other anthropometric indices of obesity”. Eating and Weight Disorders. 14:50-55.
Popkin, B.M. 1994. “The nutrition transition in low income countries: an emerging crisis”. Nutrition Reviews. 52:285-98.
Tanner, J. 1990. Fetus into Man. Cambridge: Harvard.
Tawfik-Shukor, A., Khoshnaw H. 2010 The impact of health system governance and policy processes on health services in Iraqi Kurdistan. BMC International Health and Human Rights. 10:14. Accessed September 24, 2010 at :http://www.biomedcentral.com/1472-698X/10/14.
World Health Organization. 2007. Republic of Iraq: Iraq Family Health Survey Report.
- Cohen, B.B. 2012. “Conducting evaluation in contested terrain: Challenges, methodology and approach in an American context”. Evaluation and Program Planning. 35: 189-198; Cohen, N, and Arieli, T. 2011. “Field research in conflict environments: Methodological challenges and snowball sampling”. Journal of Peace Research. 48:423-435. [↩]
- Tawfik-Shukor A., Khoshnaw H. 2010. “The impact of health system governance and policy processes on health services in Iraqi Kurdistan”. BMC International Health and Human Rights. 10:14. Accessed September 24, 2010 at :http://www.biomedcentral.com/1472-698X/10/14. [↩]
- Lohman G, Roche AF, & Martorell R. 1988. Anthropometric Standardization Reference Manual. Chicago, IL: Human Kinetics Books. [↩]
- Burger SE, Pierre-Louis JN. 2002. How to assess anemia and use the Hemocue. New York: Hellen Keller Worldwide. [↩]
- Field JO & Russell RM. 1992. “Nutrition mission to Iraq for UNICEF”. Nutrition Reviews. 50: 41-46. [↩]
- World Health Organization. 2007. Republic of Iraq: Iraq Family Health Survey Report. [↩]