In this population-based study of men and women greater than 60 years residing in Greater Manchester, UK the prevalence of clinically diagnosed Paget’s disease was 0.174%. Compared to those who were white, prevalence was greater among those identifying as Black or Black British (0.344%) and lower among those identifying as Asian or Asian British (0.048%). There was a small increase in the likelihood of disease with increasing levels of deprivation. After adjustment for other factors linked with poor COVID outcomes, those with Paget’s disease had a small though non-significant increase in the risk of admission to hospital within 28 days of a positive COVID test.

Our data are consistent with previous studies showing an increase in occurrence of clinically diagnosed Paget’s disease with increasing age and a greater incidence in men than women [4]. Our findings are also consistent with data from the UK general practice research database suggesting an increase in risk with increasing deprivation [5]. Also recent data from Quebec, Canada, using data from health administrative databases, suggesting a link with increasing social and material deprivation [16]. The explanation for this remains uncertain; selection bias seems unlikely as those living in areas of greater deprivation are, if anything, less likely to consult their primary care physician and thus to be clinically diagnosed. Supporting the view that socioeconomic factors may influence occurrence is the observation in a recent case–control study of a link between Paget’s disease and low education level[17].

Recent studies have reported a decline in the incidence of clinically diagnosed Paget’s over the past 30 years [4, 5]. Based on data obtained during 1988–1999 it was estimated (using incidence and mortality rates) that the prevalence of clinically diagnosed disease among those age 55 years and older was 0.3%. Our data provide a robust estimate of the current prevalence and suggest that 0.174% of people aged 60 years and over have clinically diagnosed disease.

How do our data compare with findings relating to the occurrence of radiographic disease. The most recent data concerning radiographic prevalence derives from a survey of 1,000 stored abdominal and pelvic CT images in Lancaster, UK [18]. Evaluation of the images suggests a radiographic prevalence of 0.8% in men and women aged 55 years and over. Comparison with our findings suggests that somewhere between one in four and one in five of those with radiographic evidence of the disease will come to clinical attention. Caution however is needed in extrapolating these data to other parts of the UK as there is important geographic variation in disease occurrence with rates highest in the NW England [1, 2].

In our study we found a higher prevalence of clinically diagnosed disease among blacks than whites. To our knowledge there are no previous data relating to the population occurrence of Paget’s disease in black people in the UK. Evidence from sub-Saharan Africa suggest a relatively low prevalence in native Africans. However, in a radiological survey in Johannesburg, South Africa Guyer reported a radiographic prevalence of 1.3% in blacks compared with a prevalence of 2.4% among whites [12]. In a survey of two cities in USA (New York and Atlanta) the prevalence of disease was found to be slightly higher in whites than blacks in New York (3.9% vs 2.6%) [11]. In Atlanta, however, interestinglyPaget’s Disease was slightly greater among blacks than whites in Atlanta (1.2% vs 0.9%), and among Atlanta men the disease was twice as frequent among blacks (1.9% vs 0.9%). In a more recent analysis of patients attending the Birmingham (USA) VA Medical Centre over a 20 year period, Paget’s disease appeared to be more common among African American than White patients (0.51% vs 0.4%) [19]. Using data from NHANES 1 the prevalence of Paget’s (based on information from pelvic radiographs) was similar in whites and blacks (0.72% vs 0.73%) [13].

Asians living in the greater Manchester area are of predominantly Indian, Pakistani and Bangladeshi origin. To our knowledge there are no radiographic survey data concerning occurrence of Paget’s in the Indian subcontinent or data concerning occurrence of Paget’s among Asians who live elsewhere. There were no cases of Paget’s reported among those of Asian background in an NHANES survey using pelvic radiographs, however, the numbers studied were small (n < 38) and their Asian origin was not specified [13]. Although traditionally considered to be uncommon in Asia, PDB has been increasingly reported from the Indian subcontinent over the last two decades though the data are primarily in the form of case reports / case series [9, 20,21,22]. In a series of 28,000 patients with diabetes, Paget’s disease was estimated in 0.066% [23].

There are reports of patients with Paget’s disease of Asian (Indian subcontinent) origin living in the UK and New Zealand though the numbers of patients is small [14, 24]. The reason for the diagnosed low prevalence among those identifying as Asians and Asian British in Greater Manchester, compared to whites or those who identify as Black or Black British is unknown. Both genetic and environmental factors are involved in the pathogenesis of Paget’s, and it is possible variation may be due to differences in one or more susceptibility factors and for which further research is needed.

In our data Paget’s disease was associated with a small though non-significant increased risk of severe COVID. Not everyone, however, who tested positive for COVID was recorded on the dataset and not all admissions during this time were due to COVID. Any misclassification due to underreporting and to non-COVID related admissions seems unlikely to be related to the occurrence of disease and would probably tend to reduce the likelihood of finding a biologic association. We were not able to adjust the findings for comorbid factors, which may have been linked with Paget’s and also adverse COVID outcome. There was evidence also of an increased risk of severe COVID (admission to hospital) linked with increasing age, gender (men > women) and increasing levels of deprivation [25].

Our data is based on population sample of people registered with their GP. There are important limitations to be considered when interpreting our findings. Classification of ethnicity was based on self-report and many people declined to define their ethnicity while others classified themselves as mixed race; the latter included those of white and Caribbean background, white and black African and white and Asian and other mixed race.

Misclassification of ethnic status may have potentially resulted in either an under- or over estimation of the true occurrence of disease among individual ethnic groups. Any such misclassification is, however, if anything to reduce the likelihood of finding significant biologic associations. As outlined our data relate to those with clinically diagnosed disease. Factors influencing clinical presentation, including for example comorbidity and health seeking behaviour may potentially impact on the likelihood of an individual being diagnosed with the disease and it is possible that such factors may explain some of the observed variation in occurrence by ethnic group. Our data concerning deprivation is derived from census data and based on current residence; and may not therefore reflect levels of deprivation experienced during the life course. Finally, our data are based on data from a large urban conurbation in the northwest of England and some caution is needed in extrapolating the findings beyond this group.

In summary, the prevalence of clinically diagnosed Paget’s disease of bone in Greater Manchester in 2020 was 0.174%. Prevalence increased with increasing deprivation and was compared to whites, more common among those who identified as Black or Black British and less common among those who identified as Asian or Asian British. Further research is required to confirm these findings and to determine whether such differences are due to variation in disease occurrence or disease presentation and also the causes of such variation.