Bell’s palsy is an acute peripheral facial nerve palsy, and about 70% of patients will have complete resolution even without treatment. The etiology is thought to be inflammatory and associated most commonly with herpes simplex virus type 1 infection ^{Reference Murakami, Mizobuchi, Nakashiro, Doi and Hato1} and, less commonly, with varicella-zoster virus infection ^{Reference Peitersen2} and a variety of other viral and nonviral infections. ^{Reference Morgan and Nathwani3} The annual incidence rate has been reported as 11.5 to 40.2 cases per 100,000 population. ^{Reference De Diego, Prim, Madero and Gavilan4} The lifetime risk has been estimated as 1 in 60. ^{Reference De Diego-Sastre, Prim-Espada and Fernandez-Garcia5} Authoritative reviews indicate that there is no racial or ethnic predilection with Bell’s palsy, ^{Reference Ronthal, Greenstein, Post, Shefner and Goddeau6,Reference Eviston, Croxson, Kennedy, Hadlock and Krishnan7} although De Diego et al. ^{Reference De Diego, Prim, Madero and Gavilan4} have raised the possibility of a racial susceptibility.

There is no epidemiological data on Bell’s palsy in First Nations populations. Among the Nêhinaw (Cree) of northern Manitoba, Bell’s palsy was known as “twisted mouth,” ^{Reference Bruyere and Garro8} and it was considered a well-known condition. During a clinical neurology practice in Manitoba and Northern Ontario extending over a 30-year period, one of us (ACJ) noted patient histories and observations of family physicians suggesting a higher-than-normal incidence and lifetime prevalence of Bell’s palsy in the First Nations populations in both provinces. In addition, bilateral, recurrent and multiple family members were affected at greater rates than expected. In an attempt to find objective evidence and quantification of these observations, a retrospective cohort study was performed to evaluate the Bell’s palsy prevalence in both First Nations and non-First Nations populations of Manitoba, Canada, with type 2 diabetes mellitus (T2DM). This was a substudy of another study evaluating the incidence and prevalence of T2DM in First Nations and non-First Nation populations, which showed greater involvement in the First Nations population. ^{Reference Ruth, McLeod, Yamamoto, Sirski, Prior and Sellers9} Hence, the focus was confined to populations with T2DM.

The Manitoba Population Research Data Repository at the Manitoba Centre for Health Policy (MCHP) contains de-identified data from multiple administrative and clinical sources, linked on a per-project basis by a scrambled personal health identification number. Manitoba has a provincially funded, universal healthcare system that covers virtually the entire population. This study was conducted as a partnership between researchers at MCHP and the First Nations Health and Social Secretariat of Manitoba to ensure that all aspects of the study included the perspectives of First Nations people and communities. Identification of registered First Nations individuals and communities was possible through the use of the Manitoba First Nations Research File. Ethics approval was received from the ethics review board of the University of Manitoba and all appropriate data custodians.

Individuals with T2DM were identified by modifying the National Diabetes Surveillance System definition to exclude non-T2DM in children and adults aged 7 years and older, and results were evaluated in 2-year blocks from 2013–2014 and 2016–2017 as previously described. ^{Reference Ruth, McLeod, Yamamoto, Sirski, Prior and Sellers9} Bell’s palsy was identified by looking back from 2016–2017 to 1979–1980 for all individuals identified in the cohorts above. Bell’s palsy was identified using the facial palsy ICD-9-CM diagnosis code 351 recorded at a physician visit. The age- and sex-adjusted percentage with one or more physician visits with a diagnosis of facial palsy was calculated by ethnicity. The relative rate of Bell’s palsy was estimated in a generalized linear model with a negative binomial distribution and the log of the person-years included as an offset. Covariates included ethnicity (First Nations vs. all other Manitobans), region of residence, biological sex and 5-year age group. The model fit well with a deviance/DF = 0.9428. In this study, lifetime prevalence refers to whether individuals have ever had one or more episodes of acute facial palsy during the available data years.

This study of T2DM in virtually the entire population of Manitoba, separated into the First Nations and non-First Nations populations, provided the opportunity to quantitatively evaluate the lifetime prevalence of Bell’s palsy in these populations. There is some uncertainty about the effect of T2DM as a risk factor for the development of Bell’s palsy. A recent study of genetic risk factors for the development of Bell’s palsy in individuals of European ancestry ^{Reference Liu, Sun, Bi, Mu, Li and Hu10} found that hypertension is a significant risk factor, but T2DM and body index mass are not. In contrast, a recent Korean study ^{Reference Seo, Ryu, Lee and Chung11} found an increased incidence rate of acute facial palsy in people with types 1 and 2 diabetes compared to those without diabetes with an incidence of 36 cases versus 25 per 10,000 person-years, respectively.

In this cohort of persons with T2DM aged 15 years and older, there were 15,115 First Nations individuals and 79,991 other Manitobans. 9.9% of the First Nations population and 3.9% of all other Manitobans with T2DM have had at least one episode of acute Bell’s palsy (crude rates). Age- and sex-adjusted rates are shown in Figure 1. Due to the much younger age of First Nations individuals in this cohort, the adjusted rates differ substantially from the crude. The adjusted relative risk was 3.1% [95% CI 2.9–3.2, p < 0.0001]. Furthermore, the increased prevalence in the First Nations population was found to be significant in all five different health regions in different geographic locations in the province. The highest prevalence was in the Northern Health Region, which was significantly higher than the Manitoba’s First Nations rate (p < 0.01). We speculate that differences in the climate and social determinants of health in this region could be important. A higher rate was also seen when point prevalence was calculated in persons currently living with T2DM during fiscal years 2013–2016. The age- and sex-adjusted rate for the First Nations population was 1.4% compared to 0.4% for all other Manitobans.

Figure 1. Prevalence of a diagnosis of facial palsy for First Nations and all other Manitobans aged 15 years and older with type 2 diabetes mellitus by health region. Age- and sex-adjusted percentage, data since 1979.

We have confirmed a significantly higher prevalence of Bell’s palsy in the First Nations population of Manitoba. Bell’s palsy was not previously recognized to occur or be more prevalent in any particular ethnic group, although studies have not specifically evaluated race or ethnicity in their study populations. De Diego et al. ^{Reference De Diego, Prim, Madero and Gavilan4} raised the possibility of racial susceptibility. In their 1999 report, they cited 16 reports of studies of the incidence of Bell’s palsy in different parts of the world. The incidence range they found was 11.5 (in Sweden) to 40.2 (in Japan) cases per 100,000 population. All four of the studies from Japan were in the upper part of the range: 27.9, 30.0, 30.5 and 40.2. This suggests the possibility of an increased incidence of Bell’s palsy in the Japanese population due to genetic or environmental factors or a combination. None of the other reports were from Asia, and conclusions about incidence in populations should not be made from incidences observed in only one or two reports.

Our results cannot be extrapolated to populations without diabetes, and diabetes may be an important risk factor for Bell’s palsy as it is in other focal neuropathies. We do not have data on the severity or on the relative rates of recovery of Bell’s palsy in this study. No information was known about the glycemic control in individuals at the time of Bell’s palsy diagnosis. We cannot exclude the possibility that worse glycemic control in the First Nations populations could have played a predisposing role for Bell’s palsy.

Our recognition of the increased prevalence of Bell’s palsy in a First Nations population may be related to genetic or environmental factors or a combination of these factors. It is also possible that differences in rates are due to other differences experienced by First Nations populations in Manitoba. These include poorer social determinants of health and the impact of racism and colonialism on access to and quality of care by the healthcare system. It would be of interest to know whether the increased prevalence is present in all five linguistic groups in First Nations populations in Manitoba. Some First Nations elders have stated their belief that we need to “examine the water” and wonder if resource extraction, its environmental effects and erosion of access to traditional foods and medicines may have an impact on disease rates. Also of interest is whether the increased prevalence is also present in First Nations populations in locations outside of Manitoba and Canada. Further epidemiological studies in potentially susceptible populations may provide insights into the pathogenesis of Bell’s palsy.