Molluscum contagiosum (MC) is a superficial infection of the dermis caused by a virus in the family Poxviridae. In otherwise healthy individuals, infection with Molluscum contagiosum virus (MCV) results in a benign self-limiting condition marked by the formation of distinctive, persistent dermal lesions that evolve slowly over the course of several weeks to several months. The total time-course of infection may be prolonged due to inadvertent autoinoculation of the virus to other parts of the body. Activities or circumstances that involve skin-to-skin contact (e.g., play, sports such as wrestling, sexual activity, etc.) have been associated with increased risk for infection [1].

Because of the characteristic appearance of MC lesions, diagnosis is generally made without laboratory testing. Often specific treatments or therapies are not pursued for MC infection in immune competent individuals, as lesions will resolve with time, however, mechanical removal (via curettage, cryotherapy, or laser treatment) and various topical therapies (including tretinoin, cantharidin, Imiquimod, cidofovir) are sometimes utilized to minimize the duration that lesions are present, particularly on the face or on areas of the body that are subject to heightened irritation. Molluscum contagiosum in persons who have immune compromise—whether due to HIV infection, immunosuppressive drug therapies, or other reasons—can be complicated.

Persons with atopic dermatitis (i.e., eczema) are generally considered to be at greater risk for contracting MC due to immune deficits, compromised skin integrity, or both [2]–[4]. However, investigations into the association between atopic dermatitis and molluscum infection have been observational in nature and have principally addressed the conditions only in children. In general, epidemiologic studies of MC are few and most have focused on rates of infection or risk factors for infection in specific population subgroups. These subgroups have been defined variously by geographic location [5], [6] subject age [7], [8], immunologic status [9], [10] or the type of medical service facility at which they had sought care (e.g., tertiary care facilities, dermatology specialists, etc.) [7], [11]. Provision of population-based incidence rates for MC, which would allow for comparison across studies, is rare.

With this study we seek to provide an extensive description of the epidemiology of MC among American Indian and Alaska Native (AI/AN) persons, a subset of the general US population. The AI/AN population is not anticipated to directly reflect the broader US population with respect to health issues – AI/AN communities are often subject to considerable health disparities—but approximately 57% of the AI/AN population across the United States is served by the Indian Health Service (IHS); therefore utilization of healthcare services through the IHS can provide a broadly encompassing, population-based, view of specific health-related issues including infection with MCV. The IHS National Patient Information Reporting System (NPIRS), which records 100% of clinic and ambulatory care records for all IHS-contracted or operated facilities, captures population-based outpatient visit data through time from across the United States for all AI/AN age groups. This study, which spans five years, represents the largest, most comprehensive molluscum survey performed to date. The MC-associated outpatient visit rate for AI/AN persons is compared with that of the general US population (derived from data available through the National Ambulatory Medical Care Survey [NAMCS] and the National Hospital Ambulatory Medical Care Survey [NHAMCS]). Patient-based MC-associated outpatient visit rates or incidence among AI/AN persons are examined.

The IHS direct and contract outpatient visit data for AI/AN persons for calendar years 2001–2005, obtained from the IHS National Patient Information Reporting System (NIPRS), were analyzed [12]. The direct and contract health service outpatient visit data consist of all direct and contract health care outpatient visit records from IHS-and tribally-operated facilities and facilities which are contracted with the IHS or with specific tribes to provide healthcare services to eligible AI/AN persons [12], [13]. All AI/AN persons in this study received IHS-funded health care and may not be representative of all AI/AN persons in the United States.

Molluscum contagiosum-associated outpatient visit records, where MC was listed as one of up to nine diagnoses, were selected for analysis. Visits with MC listed as one of up to three diagnoses were selected from the US ambulatory care data. The International Classification of Diseases, 9^th Revision, Clinical Modification (ICD-9-CM) code 078.0 was used to define MC [14]. The IHS provided outpatient visits by patient (no unique identifiers were provided to investigators), which enabled the examination of multiple outpatient visits for patients and estimation of the incidence of MC. The patient's first annual MC visit during each year of the study period was used for patient-based analysis to estimate the incidence, reported as number of AI/AN persons with at least one MC-associated outpatient visit per 10,000 AI/AN persons. Because this study only encompasses individuals who have sought healthcare while infected with MCV, it is likely to be a minimum estimate of the true incidence of MC in the AI/AN population. The primary unit of the analysis in this study was the patient, unless otherwise stated.

Both molluscum contagiosum-associated outpatient visits and patient-based outpatient visits for 2001–2005 among AI/AN persons were examined by age group (<1 year [infants],1–4, 5–14, 15–19, 20–44, 45–64 and ≥65 years), IHS region (West Northern Plains [Aberdeen and Billings areas], East Northern Plains [Bemidji area], Southern Plains [Oklahoma area], Southwest [Albuquerque, Navajo, Phoenix and Tucson areas], East [Nashville area], Alaska, and West [California and Portland areas]), sex and etiology. Month of visit and diagnoses listed with a MC diagnosis on the record were examined. Annual outpatient visit rates and incidence rates were calculated and expressed as the number of outpatient visits and patients per 10,000 AI/AN persons, respectively. The IHS user population estimates of AI/AN persons within the IHS healthcare system for fiscal years 2001–2005 were used to calculate the rates [13], [15]. The user population is all AI/AN persons who were eligible to receive IHS-funded healthcare service during the 5-year study period. Rate comparisons among AI/AN persons were evaluated by characteristic using risk ratios (RRs) with 95% confidence intervals (CIs) [16]. Comparisons of the interval in days between outpatient clinical visits by age group were made using the Wilcoxon rank-sum test.

The overall Molluscum contagiosum-associated average annual outpatient visit rates were examined for the general US population. NAMCS and NHAMCS for 2001–2005 were used to examine outpatient visits [17], [18]. The NAMCS and NHAMCS provide data with respect to outpatient visits and provide an assessment of outpatient visits for the general US population. The basic sampling unit for the NAMCS is the physician-encounter. Only visits to the offices of non-federally employed physicians classified by the American Medical Association or the American Osteopathic Association as “office-based, patient care” are included. The NHAMCS comprises a national probability sample of visits to the emergency and outpatient department of non-institutional general and short-stay hospitals. The US national NAMCS and NHAMCS data in this study do not include IHS/tribal facility outpatient visits or identify multiple hospitalizations by patient. For the general US population, the outpatient visit rate was expressed per 10,000 persons using the US Census estimates [19]. Standard errors of NAMCS and NHAMCS estimates were calculated by using SUDAAN software to account for the sampling designs [20], and used to calculate 95% CIs for the rates [16]. If the relative SE of estimates exceeded 0.30 or if the number of unweighted visits in a strata was <30, the estimates were considered unreliable and are not reported [17], [18].

Despite the ubiquity of MCV throughout the world, few recent studies have addressed incidence trends or burden of infection associated with this virus, and none that we know of have provided population-based rates to describe incidence or health care utilization. The purpose of this study has been to provide a comprehensive view of epidemiology of MC across different age groups and in different parts of the United States, by utilizing the health service information for AI/AN persons available through the IHS, which affords a more detailed accounting of outpatient service utilization than is available for the general US population.

Our findings point to a relatively high overall MC-associated outpatient visit rate in the AI/AN population (20.15/10,000 AI/AN persons). This rate is comparable to that estimated for the US population overall, although the rate for AI/AN persons living in the West, East and Alaska regions may be higher than that for the general US population.

A national survey of MC-associated physician visits performed over two decades ago, using data obtained from the National Disease Therapeutic Index Survey (NDTIS) of the United States demonstrated an increase in the proportion of MC-associated clinic visits (per total clinic visits) during a 18-year period from 1966 to 1983, ranging from 1.2 to 11 MC-associated visits per 100,000 clinic visits [21]. In contrast, here, we saw relatively high but stable incidence rates over the 5-year period of observation. It is important to note, however, that ‘clinic visits’ were used as the denominator for MC-associated health care utilization rates in the NDTIS study, whereas in this study we report annual MC-associated outpatient incidence per 10,000 AI/AN persons, therefore rates are not directly comparable.

In addition to high overall incidence, we note relatively high incidence among specific population sub-groups, notably 1–4 year olds in the East and West regions where incidence rates of 123.62/10,000 and 150.08/10,000, respectively, were reported. These incidence rates are comparable to that for chickenpox prior to initiation of vaccination (1980–1990), which was estimated to be 149.8/10,000 [22]. Recent studies which measured age-related health care utilization for MC concluded that rates of health care utilization, and overall numbers of MC-associated visits were highest among children 2–5 years old [7], [23]. Formerly, onset of schooling was typically viewed as a social risk factor for acquisition of MC, particularly in developed, affluent nations [1]; a study performed in Holland in the late 1980s provided evidence consistent with this idea [11]. However, increased utilization of child day care in the United States during the last two decades has generally lowered the age of socialization (outside the immediate household) among children across most socio-economic strata [24]. It bears further consideration whether the observed elevated burden of infection observed among pre school-aged children in this study reflects general tendencies toward communal care of very young children in the AI/AN population, or possibly heightened utilization of child daycare.

Independent of age, both poverty and crowded living quarters have been hypothesized as general risk factors to account for perceived higher rates of MCV particularly in the tropics [1], [8]. Moisture and humidity have been implicated as relevant climatic factors as well [1], [5], [25], but the strength of association between poverty, geography or climate, and incidence of infection with MCV remains undetermined. In our study, overall incidence of MC-associated clinic visits was observed to be highest in the West, East and Alaska regions, and lowest in the Southwest and Southern Plains regions, the latter of which are arguably the areas with warmest climate. These regional differences in incidence were stable over the 5-year period of observation; it seems unlikely that the observed differences could be attributable to climate alone.

For individual AI/AN patients, however, the presence of concurrent skin conditions was often attendant MC-associated visits. For purposes of this study, these conditions were broadly categorized into those associated with immune deficits with dermatologic manifestations, for example atopic dermatitis and eczema, versus those suggestive of breaches in skin integrity, such as acne and impetigo. Conditions in either category could predispose persons to MCV infection (eczema is a risk factor for severe complications from infection with poxviruses and other cutaneous viral pathogens) but, notably, the relative frequencies of each category of skin condition varied in line with MC patient age. Atopic dermatoses were diagnosed concurrently with MC in approximately 8% of all persons under the age of 14, and in ∼18% of persons ≥65 years of age, but these diagnoses were relatively infrequent among persons 15–64 years old. Conversely, among 15–64 year olds, a greater proportion of patients were diagnosed with acne or erythematos conditions, both associated with loss of skin integrity. These observations suggest that risk reduction messages tailored for AI/AN patients of specific ages could be useful in preventing MCV infections. (For example, teenagers with acne should be enjoined to take special care when interacting with a younger sibling that has MCV infection.)

This study demonstrates that within the AI/AN population in the United States there is appreciable utilization of health care services for MCV infection, but there appears to be a relative absence of use of specific medical interventions which could be employed to diminish the consequences of infection (although prescription treatments were not specifically assessed). Pediatric and adult treatment options include curettage and application of various topical agents [26], [27]. The risk of MCV transmission and subsequent infection (either to oneself or to others) can be lessened by successful use of interventions that hasten the resolution of lesions. But physicians need always to balance the potential value of the intervention against the potential risks, i.e., scarring, pain, etc. [26].

The IHS/tribal and the general US population outpatient visit data are very useful for evaluation, although there are some limitations to the data. Only AI/AN persons who had received medical evaluation for MC are included in the estimate in this study, therefore the incidence rate reported here may be an underestimate of the true incidence rate of MC occurrence in the AI/AN population, and regional variations in health-care seeking behavior may exist. Also, diagnoses may be incomplete, miscoded or vary by region. The AI/AN user population is an estimate of the number of AI/AN persons who are eligible to utilize the IHS healthcare system and may not include all AI/AN persons who seek medical consultation. Further, the AI/AN persons in this study are not representative of all AI/AN persons in the United States. For the general US population, only the overall MC outpatient visit rates were estimated due to the small number of MC visits in the surveys. Also, multiple outpatient visits for individual patients are not identified; therefore, the incidence of MC can not be estimated for the general US population.

In sum, this study highlights the need for periodic population-based measurements to assess long-term trends in incidence and healthcare utilization for MC, particularly as regards the age of initial MCV acquisition, associations with other concurrent skin conditions, and availability and use of medical interventions. Though this study specifically addresses MC-associated health care utilization in the AI/AN population in the United States, our results highlight features of MC epidemiology that merit further consideration and evaluation for other populations in the United States and elsewhere. Because the need seems great, the AI/AN population should benefit from community outreach regarding MC prevention, particularly in very young children.