Table 4 Characteristics of included publications: Diabetes Mellitus

Abukhdeir (2013) [32]

Palestinian Territories: West Bank/Gaza

EMRO

Cross sectional

Palestinian households in the West Bank and Gaza Strip

May 2004—July 2004

4456 households in the West Bank and 2118 in the Gaza Strip

Being a refugee was a significant risk factor for diabetes and CVD while being married/engaged or divorced/separated/ widowed was a risk factor for diabetes and hypertension. Non-refugees were 33% less likely to have diabetes and 46% less likely to have CVD than refugees. Gender was a risk factor for hypertension with females being 60% more likely to have hypertension than males

Ahmad (2015) [33]

Syria

EMRO

Situational analysis using document analysis, key informant interviews, and direct clinic observation

Syrian national health system

October 2009 -August 2010

53 semi-structured interviews

The rebuilding of a post-conflict heath care system in Syria may benefit from insights into the structural problems of the pre-crisis system. Weaknesses that existed before the crisis are compounded by the current conflict. The authors suggest an over reliance on secondary and tertiary care for DM patients with withdrawal of the Syrian government from the public health clinics, which led to escalating healthcare costs and fostered increasingly unequal access

Alabed (2014) [79]

Country of Asylum: Syria

Country of Origin: Palestinian Territories

EMRO

Cross sectional

Palestinian refugees living in Damascus attending three UNRWA health clinics

August 2008—September 2008

154 DM patients

UNRWA clinic inspections highlighted shortages in drug stocks with 47.3% of patients reporting problems accessing prescribed medications and 67.7% reporting having to buy medications at their own expense at least once since their diagnosis. Patients’ knowledge of their condition was limited, Patients were generally unaware of the importance of good glucose control and disease management. Women were more likely to attend the clinic than men, with 71% of patients being female

Ali-Shtayeh (2012) [80]

Palestinian Territories: West Bank

EMRO

Cross sectional

Patients attending outpatient departments at West Bank Governmental Hospitals in 7 towns in the Palestinian territories (Jenin, Nablus, Tulkarm, Qalqilia, Tubas, Ramalla, and Hebron)

August 2010—May 2011

1,883 DM patients

While all patients using complementary and alternative medicine (CAM) were additionally using conventional therapies, the use of CAM differed significantly between residents of refugee camps versus residents of urban or rural areas (p = 0.034). More residents in a refugee camp reported using CAM vs. not using CAM as compared to those who reported living in a village or city. Most CAM users were above 40 years old, predominantly female, and residents of refugee camps and rural areas

AlKasseh (2013) [81]

Palestinian Territories: Gaza

EMRO

Retrospective case control

Refugee women attending the UNRWA postnatal clinics in Gaza

March 2011—June 2011

189 postnatal GDM women with 189 matched controls by age and place of residency

A history of miscarriage more than once, being overweight before pregnancy, history of stillbirth, history of caesarean birth and positive family history of diabetes mellitus were strongly correlated with developing gestational diabetes(GDM). WHO criteria for screening for GDM remain a good instrument to identify GDM in refugee populations in war-torn countries (like the Gaza Strip)

An (2014) [82]

China

Western Pacific

Retrospective cohort

1976 Tangshan Earthquake survivors, aged 37–60, without severe liver disease, trauma surgery, secondary diabetes, or diagnosed mental disease

September 2013—December 2013

1030 exposed subjects

The incidences of impaired fasting glucose and DM for earthquake survivors were significantly higher than that for the control group. There was a higher diabetes incidence in those who had lost relatives than those who had not lost relatives, however, this effect was only statistically significant in women earthquake survivors

Armenian (1998) [83]

Armenia

Europe

Retrospective cohort

Employees of the Armenian Ministry of Health and their immediate families who survived the 1988 Earthquake in Armenia

1990–1992

35,043 persons (7,721 employees who had survived the disaster and their family members)

Longer term increased rates of DM morbidity following an earthquake are related in a dose–response type relationship to the intensity of exposure to disaster. Bereavement, injuries in the family, and material loss, act as independent predictors of long term adverse physical illness including for DM

Balabanova (2009) [84]

Georgia

Europe

Rapid appraisal process with snowball sampling

Georgian health system evaluation

March—April 2006

36 interviews

Essential inputs for diabetes care are in place (free insulin, training for primary care physicians, financed package of care), but constraints within the system hamper the delivery of accessible and affordable care. The scope of work of primary care practitioners is limited and they rarely diagnose and manage diabetes, which instead takes place in the context of a hospital admission and tertiary-level endocrinologists. Obtaining syringes, supplies and hypoglycemic drugs and self-monitoring equipment remains difficult and leads to a cost driven shift toward insulin for diabetic management

Ben Romdhane (2015) [85]

Tunisia

EMRO

Situational analysis

Tunisian national health system

2010

12 key informants were interviewed and eight documents were reviewed

Weaknesses that existed before the 2011 Revolution(Arab Spring) were compounded during the revolution. This study was conducted prior to political conflict but written post-conflict. Growth of the private sector fostered unequal access by socioeconomic status and reduced coordination and preparedness of the health system

Besancon (2015) [86]

Mali

Africa

Case study

Mali diabetic population following a March 2012 Coup in Bamako

Spring 2012 following the March 2012 coup

–

Diabetics are a vulnerable population in humanitarian crisis due to their continuous need for health care and medicines and the financial burden this may place on them. The authors propose that in an emergency setting there is not one single diabetes population that should be considered in planning humanitarian responses, but multiple, each with unique needs. These sub-populations include people still in active conflict regions, IDPs, refugees, and the population which houses IDPs

Ebling (2007) [87]

Croatia

Europe

Multipart study including both a retrospective cohort study and an uncontrolled before-after study

Refugee-returnees of the 1991–1992 war operations in Eastern Slavonia from Osjek-Baranga County, Croatia

2003

retrospective cohort study: 589 participants

uncontrolled before-after study 202 participants

The participation of subjects with DM in the population of refugee-returnees despite similar demographic indicators, exceeded values for both Slavonia and Croatia. Extremely high participation of patients with diabetes was noted(10.5%), despite a lower proportion of aged people over 65 among returnees

Eljedi (2006) [88]

Palestinian Territories: Gaza

EMRO

Cross sectional

Patients with DM residing in refugee camps in Gaza Strip

November 2003—December 2004

197 DM patients

Using the World Health Organization Quality of Life questionnaire (WHOQOL-BREF) four domains–including physical health, psychological, social relations, and environment – were strongly reduced in diabetic patients as compared to controls, with stronger effects in physical health (36.7 vs. 75.9 points of the 0–100 score) and psychological domains (34.8 vs. 70.0) and weaker effects in social relationships (52.4 vs. 71.4) and environment domains (23.4 vs. 36.2). The impact of diabetes on health-related quality of life (HRQOL). was especially severe among females and older subjects (above 50 years)

Gilder (2014) [89]

Country of Asylum: Thailand

Country of Origin: Myanmar

South-East Asia

Cross sectional

Women attending the antenatal care (ANC) clinic in Maela refugee camp on the Thai–Myanmar border

July 2011—March 2012

228 women

The prevalence of GDM is lower in this population compared with other populations, but still complicates 10% of pregnancies. Despite the weight of evidence for the benefits of early diagnosis and treatment of GDM, the absence of a simple, inexpensive and applicable screening method remains a major barrier to GDM screening programs in refugee camps and other resource-poor settings

Habtu (1999) [90]

Ethiopia

Africa

Cross Sectional

Insulin treated diabetic patients from the Diabetic Clinic at the Mekelle Hospital in rural Tigray, Northern Ethiopia- the center of the severe Ethiopian famine of the mid-1980s

Six month period in 1997

100 patients

The correct prescribed dose of insulin was only being taken by 50% of patients and the correct syringe by only 12%. Insulin treatment had been interrupted in 48% of cases due to lack of supply. Low BMI(mean of 15.8), young age, and resistance to diabetic ketoacidosis(DKA) amongst study participants were consistent with previous descriptions of malnutrition related diabetes mellitus(MRDM)

Hult (2010) [42]

Nigeria

Africa

Retrospective Cohort

40 year old Nigerians with fetal exposure to famine in Biafra, Nigeria during the Nigerian civil war (1967–1970)

June 2009–July 2009

1,339 study participants

Fetal and infant undernutrition was associated with significantly increased risk of impaired glucose tolerance in 40 year old Nigerians. However, early childhood exposure was not associated with increased risk

Kallab (2015) [44]

Country of Asylum: Lebanon

Country of Origin: Syria

EMRO

Program implementation reflection

Syrian refugees and vulnerable Lebanese host communities over the age of 40

November 2014- May 2015

1825 patients

DM accounted for 54% of patient cases, with 27% of patients affected by both DM and HTN. Principal barriers to providing diabetic management in active conflict included insecurity, the fluid movement of refugees, limited opening hours of the centers, transportation costs, and medication shortages

Karrouri (2014) [91]

Country of Asylum: Tunisia

Country of Origin: Libya

EMRO

Case report

Case of a 10-year-old Libyan boy

–

One patient

Report of a 10 year old without personal or familial diabetes mellitus history who developed type 1 diabetes appeared immediately following severe psychological trauma

Khader (2012) [82]

Country of Asylum: Jordan

Country of Origin: Palestinian Territories

EMRO

Retrospective cohort

Persons with DM at Nuzha PHC Clinic

October 2009- March 2012

2,851 patients

A directly observed therapy(DOTS) cohort monitoring system can be successfully adapted and used to monitor and report on Palestinian refugees with DM in Jordan. A sizeable proportion of DM patients of the clinic failed to have postprandial blood glucose measurements, and BP measurements in those with comorbid HTN

Khader (2013) [77]

Country of Asylum: Jordan

Country of Origin: Palestinian Territories

EMRO

Retrospective cohort

Palestine refugees living in Jordan

October 2009- June 2013

12,549 total patients

High burden of disease due to DM amongst Palestinian refugees at UNRWA primary health care clinics in Jordan. Cohort analysis using e-Health is a successful tool for to assess management and follow-up of DM patients. Complications, including myocardial infarction and end-stage renal disease were significantly more common in males. Females were more likely to be obese

Khader (2014) [69]

Country of Asylum: Jordan

Country of Origin: Palestinian Territories

EMRO

Retrospective cohort

Palestinian refugees living in Jordan with DM attending Nunzha Clinic

2012

2,974 DM patients

E-Health systems are useful for monitoring patients, since over half who miss their quarterly appointment fail to return. Suggests a need for monitoring and active follow-up

Khader (2014) [45]

Country of Asylum: Jordan

Country of Origin: Palestinian Territories

EMRO

Retrospective cohort

Palestinian refugees living in Jordan with DM attending Nunzha Clinic

2010–2013

119 DM patients

E-health systems are useful for monitoring patients. An increasing number of patients had complications despite no change in obesity rates indicating places where more resources may be useful

Li (2010) [72]

China

Western Pacific

Retrospective cohort

Rural Chinese exposed to the Chinese famine(1959–1961) during fetal life and early childhood

2002

7,874 rural Chinese

In severely affected famine areas, fetal-exposed adults had an increased risk of hyperglycemia compared with nonexposed subjects. Differences were not significant for the early and mid childhood–exposed cohorts. This association appears to be exacerbated by a nutritionally rich environment in later life

Lumey (2015) [92]

Ukraine

Europe

Retrospective cohort

Individuals exposed to the man-made Ukrainian famine of 1932–33 during prenatal development compared with all patients with type 2 diabetes diagnosed at age 40 years or older in the Ukraine national diabetes register 2000–08

2000–2008

43,150 patients with diabetes and 1,421,024 controls

Demonstrates a dose–response relationship between famine severity during prenatal development and odds of type 2 diabetes in later life. The associations between type 2 diabetes and famine around the time of birth were similar in men and women

Mansour (2008) [93]

Iraq

EMRO

Cross sectional

Diabetic patients in an outpatient clinic in Al-Faiha general hospital in Basrah, South Iraq

January 2007—December 2007

3,522 diabetic patients

The most common reasons for poor glycemic control(HBA1C > 7%) listed by patients were drug shortages and drugs and/or laboratory expense(over 50%). 30% of diabetic patient with poor glycemic control believed that their poor glycemic control is due to migration after the war

Mateen (2012) [48]

Country of Asylum: Jordan

Country of Origin: Iraq

EMRO

Cross sectional

Iraqi refugees receiving health assistance in Jordan as recorded by a UNHCR database

January 2010-December 2010

7642 Iraqi refugees

11% of refugees presented with type 2 DM. For all refugees the largest number of visits were for essential hypertension (2067 visits); visual disturbances (1129); type II diabetes mellitus (1021)

Mousa (2010) [50]

Country of Asylum: Jordan, Syria, Lebanon, Gaza, West Bank

Country of Origin: Palestinian Territories

EMRO

Cross sectional

UNRWA registered Palestinian refugees attending UNRWA clinics

June 2007

7,762 refugees

Overall 9.8% of screened refugees had random blood glucose values ≥ 126 mg/dL. Being older than 40 years, obese or with a positive family history of diabetes or cardiovascular disease increased the risk of presenting with hyperglycemia 3.5, 1.6 and 1.2 times respectively. Variations were statistically significant between UNRWA locations and between the sexes. Significant variations were found between fields for females (χ2 = 112.6, P < 0.01) and for males (χ2 = 39.2, P < 0.01), with the highest proportion of cases diagnosed in the Occupied Palestinian Territories and the lowest in Jordan and Syria

Ramachandran (2006) [94]

India

South-East Asia

Retrospective cohort

Tsunami affected population of Chennai(Madras) in Southern India

April 2005- June 2005

1,184 tsunami affected subjects, 1,176 controls

Undetected diabetes and impaired glucose tolerance were higher in the tsunami-hit area as compared to controls. Diabetes prevalence was found to be similar in the tsunami affected population and control. Women of both the control and the tsunami affected population had both a higher stress score(using the Harvard trauma questionnaire) than men with a significantly higher stress score in women affected by the tsunami, as well as a higher prevalence of impaired glucose tolerance in the tsunami hit area

Read (2015) [67]

The Philippines

Western Pacific

Cross sectional

Patients treated by an Australian Government deployed surgical team in a field hospital in the city of Tacloban for 4 weeks after Typhoon Haiyan

November 2013

131 persons

Sepsis from foot injuries in diabetic patients constituted an unexpected majority of the workload of a foreign collaborative surgical medical team in Tacloban in the aftermath of Typhoon Haiyan

Sengul (2004) [95]

Turkey

Europe

Prospective cohort

Type 1 Diabetic Survivors of the 1999 Marmara Earthquake

1998–2000

88 subjects

HbA1c levels and insulin requirements significantly increased at the 3rd month post earthquake however only increased insulin requirement continued to be significantly increased, one year post earthquake. No significant difference was identified between HbA1c levels pre earthquake and post 1 year earthquake. Results indicated that the Marmara earthquake affected glycemic control of people with type 1 diabetes in the short term but its negative impact did not continue in long term

Sofeh (2004) [94]

Country of Asylum: Pakistan

Country of Origin: Afghanistan

EMRO

Cross sectional

Adult Afghan Refugees attending Red Cross health care facilities in Peshawar, Pakistan

.

456 patients

The frequency of non-insulin dependent DM was found to be 55.9% amongst Afghan refugees in Peshawar during a two year study period. 17.25% of diabetics had concomitant hyperlipidemia. Gender was not identified as a risk factor for higher fasting blood glucose levels

Strong (2015) [53]

Country of Asylum: Lebanon

Country of Origin: Syria

EMRO

Cross sectional

Syrian refugees over age 60 residing in Lebanon and registered with either Caritas Lebanon Migrant Center (CLMC) or the Palestinian Women’s Humanitarian Organization (PALWHO)

March 2011—March 2013

210 refugees

47% of older refugees had DM. The number of days older refugees reporting eating bread only and nothing else corresponded to their reported financial status. Financial difficulties were given as the primary reason for not seeking care by 79% of older refugees with only 1.5% stating they had no difficulties in obtaining care when needed

Wagner (2016) [96]

Cambodia

Western Pacific

Uncontrolled before and after

Unpaid Cambodian village health guide volunteers were trained in DM prevention teaching behaviors

.

185 guides were trained to instruct at 10 health centers

Knowledge of community health workers on DM prevention techniques increased significantly from pre-test to posttest after 6 months of follow-up. 159 guides (85%) completed at least one monthly checklist

Yaghi (2012) [97]

Lebanon

EMRO

Cross sectional

Cases of amputations in Lebanon

January 2007-December 2007

661 amputations

Diabetes and vascular indications were not only more common than trauma-related amputation, but both were associated with more major surgery and longer hospital stay including conflict afflicted southern Lebanon where trauma, diabetes and vascular disease amputations all occurred at more than twice the national rate

Yusef (2000)[22]

Country of Asylum: Lebanon

Country of Origin: Palestinian Territories

EMRO

Cross sectional

Diabetic and hypertensive patients attending UNRWA primary health care facilities in Lebanon

1997

2,202 records

Presence of both DM and HTN increased the risk for late-stage complications. Only 18.2% of diabetic patients and only 17.7% of DM patients with HTN were managed by lifestyle modification. About 50% of type 2 and 66% of type 1 patients who were on insulin were well controlled