Cutaneous leishmaniasis (CL) is a neglected tropical disease for which approximately 500,000 to 1,000,000 new cases are reported per year worldwide [1,2]. Furthermore, it causes an estimated 2.4 million disability-adjusted life years, placing it among the top 10 in a global analysis of infectious diseases [3]. Countries in the Eastern Mediterranean region contribute approximately 57% of the total CL burden, where Leishmania tropica and Leishmania major are endemic in 18 countries and territories (including Iraq). Moreover, more than 100,000 new cases of CL are reported annually to the World Health Organization by countries in the Eastern Mediterranean region; however, the actual incidence is estimated to be 3 to 5 times higher [1,4,5]. In Iraq, surveillance data after the 1970s showed an average of 10x00 cases per year [6]. According to internal technical reports released by the Iraqi Ministry of Health, the last country-wide outbreak started at the end of 2014 and continued throughout 2017, when the number of cases per year reached an average of 16,000. In 2018, the number of cases started to decline steadily and reached approximately 11,000.

There are more than 20 Leishmania species that can be transmitted to humans, and more than 90 sand fly species that can transmit the protozoa to humans; moreover, approximately 70 animal species, including humans, are natural reservoir hosts of Leishmania parasites [7]. The transmission cycle of the parasite in nature can be either zoonotic or anthroponotic [8,9]. In Iraq, data are lacking regarding the most common Leishmania species, reservoirs, and vectors. However, evidence from nearby countries suggests that both transmission cycles of CL (zoonotic and anthroponotic) are common in Iraq [5,10,11].

Risk factors for developing CL include residence in rural areas, climate changes, movement of people, conflict areas, deforestation, house characteristics, and human behavior [9,12-14]. Prevention and control of leishmaniasis requires a combination of intervention strategies because transmission occurs in a complex biological system involving the human host, parasite, sand fly vector, and, in some cases, an animal reservoir host. Key strategies for prevention are early diagnosis and effective case management, vector control, effective disease surveillance, control of animal reservoir hosts, and social mobilization and strengthening partnerships among all concerned institutions [14].

Although CL is a self-healing disease, it is potentially disfiguring [1]. The only drug licensed by the Iraqi Ministry of Health to treat CL is sodium stibogluconate, a pentavalent antimony compound.

The recent outbreak affected most Iraqi provinces variably, with an overall incidence rate of 0.9/10³ population. The highest incidence rate was in Diyala Province (4/10³ population), while the lowest incidence rate was in Duhok Province (0.01/10³ population). According to internal reports and discussion with the zoonotic diseases section at the Iraq Communicable Diseases Control Center, the lack of infrastructure and municipal services, the presence of hard-to-reach areas, and a lack of prevention programs were blamed for the occurrence of the outbreak. Diyala was subjected to terrorist and military operations from 2014 to 2016, when most of its residents were displaced. Meanwhile, it also encountered a wave of a Leishmania epidemic that started in November 2014, reached its peak during 2015, and continued throughout 2017. In response to the rapid escalation of the outbreak, the outbreak response team investigated the outbreak to identify possible risk factors and the impact of removing these factors on reducing the number of cases.

This is a population-based unmatched case-control study. A case of CL was defined as any person who showed clinical signs (skin or mucosal lesions) and was diagnosed by a dermatologist with CL. A control person was defined as any person (or family member) who was proved to be free of these skin or mucosal lesions. Controls were chosen preferentially according to the site where the cases occurred (from the neighboring house or village). The study was conducted in two conveniently selected districts in Diyala Province (Al-Muqdadiya and Al-Mansuriya). Those two districts were selected as part of the on-job outbreak investigation because surveillance data detected an increase in the number of CL cases in these areas, and those areas were in the recovery process after security instability. Approval for conducting the study was obtained from the Public Health Directorate/Ministry of Health and Diyala Directorate of Health. Oral consent was obtained from the cases and controls themselves or from their caretakers.

Field epidemiology training program students interviewed cases and controls using a modified questionnaire of the case investigation form of the zoonotic section of the Iraq Communicable Diseases Control Center. The questionnaire contained questions about the main demographic (age, sex, occupation), clinical (date of onset, signs and symptoms, presence of other cases within the family, treatment, previous visits, number and site of skin lesions), and epidemiological characteristics (displacement history, house and residency data [information about the type of residency area; house construction materials, such as wall type; electricity provided; animals living within the house; painting of indoor areas; presence of rodents inside or around the house]), sleeping habits, and preventive measures implemented in the area (fogging and use of bed nets).

A total of 866 persons were interviewed within the 717 families visited: 451 cases (292 from Al-Mansuriya District and 159 from Al-Muqdadiya District) and 415 controls (182 Al-Mansuriya District and 233 from Al-Muqdadiya District). However, we excluded 22 persons from the sample due to incomplete information. The final sample size used was 844 persons (cases=432, controls=412), with a ratio of almost 1 case to 1 control.

Univariate analysis was used to describe the study sample. Bivariate analysis was used to detect possible associations between each of the risk factors and the disease (CL) using the chi-square test of independence. The unadjusted odds ratio (OR) and 95% confidence interval of each risk factor were calculated using binary logistic regression. The attributable fractions and their corresponding 95% CIs were calculated for the modifiable risk factors. A P value <.05 was considered statistically significant.

Epi Info, version 7.2 was used for data entry and SPSS, version 25 (IBM Corporation) was used for data analysis.

Data from 844 persons (432 cases, 51.2%) were analyzed. There were no gender differences between cases and controls. Cases were more likely than controls to report a history of previous displacement, electricity supply for less than 12 hours per day, and living in a rural area. Regarding house characteristics, cases were more likely than controls to report living in a clay house, living in unpainted indoor areas, and the presence of rodents inside the house. As for animal ownership and the distances of the animals from the house, cases were more likely than controls to have chickens only, sheep only, or both and a mixture of animals (dogs and sheep or dogs and chickens) within a distance of less than 100 meters. Regarding possible preventive measures, cases were more likely to report fogging, bed net use, and sleeping outside or a mixture of inside and outside than controls.

Almost all the risk factors were statistically significantly associated with higher odds of having CL. Nevertheless, the strength of the association varied, as it was stronger (4 to 5 times higher odds of having CL) for factors such as displacement, having animals within 100 meters of the house, and sleeping outside the house. Factors that were associated with a 2 to 3 times increase in the odds of having CL included living in a clay house, having an unpainted indoor area, sleeping in a mixed pattern (inside and outside the house), having animals (whether chickens only, sheep only, or both, or mixtures of dogs and sheep or dogs and chickens), and, interestingly, using a bed net and fogging/unknown fogging status. In fact, the use of a bed net was associated with 72% higher odds of having CL in comparison to the lack of use of a bed net (OR 1.72, 95% CI 1.08-2.72). Likewise, fogging and unknown fogging status were associated with statistically significant 2-fold higher odds of having CL compared to no fogging (P<.001).

Regarding the impact of removing modifiable risk factors, our results show that approximately 70% to 80% of the cases were associated with displacement, the presence of rodents inside the house, the presence of animals within 100 meters of the house, the presence of animals (whether chicken only/sheep only/both or a mixture of dogs and sheep or dogs and chickens), and sleeping outside. Similarly, approximately 40% to 50% of the exposure of the cases was associated with living in a clay house; living in a rural area; having an unpainted indoor space; having an electricity supply for less than 12 hours per day; and, interestingly, using a bed net. Unexpectedly, approximately 10% to 20% of the exposed cases reported fogging or unknown fogging status. That is, fogging and unknown fogging status were negatively associated with the occurrence of CL.

The characteristics of the study sample are shown in Table 1. The risk factors for CL in the study population are summarized in Table 2.