The OI-NMQ was a feasible instrument to assess MSK pain and consequences in persons with OI. The prevalence of MSK pain among persons with OI was generally high in our pilot study, most commonly in the upper and lower back, shoulder, and hip. A larger proportion of participants with moderate to severe OI experienced MSK pain compared to persons with mild OI. The consumption of opioids among persons with moderate to severe OI was higher than in milder OI. MSK pain impacted regular work and daily activities in half of the participants. The OI-NMQ proved to be easily understandable and feasible to use in adults with OI. This assumption is based on the low number of missing answers, few comments related to the questionnaire, and no inconsistencies in the answers related to 7 days and 12 months prevalence of MSK pain. This is in concordance with other studies describing validity and reliability of the NMQ that we have based our questionnaire on [19].

Our study demonstrated a high prevalence of self-reported MSK pain among both groups of adults with OI. MSK pain can be associated with long bone fractures, skeletal deformities, and vertebral fractures in OI [9]. However, only six participants reported one or more fractures during the past year prior to participation and the fracture rate did not correlate to the large proportion of participants experiencing MSK pain during the past 12 months in our study. Thus, indicating that pain can be unrelated to recent fractures in adults with OI. Dove et al. [10] came to the same conclusion that people with OI despite the reporting of new fractures in clinic, the incidence of persistent pain could not be fully explained by current fractures. It is strongly indicating that not all MSK pain in adults with OI is fracture related.

Our questionnaire did not evaluate risk factors, or reasons, for chronic MSK pain in our participants. The high proportion of MSK pain in our study could be caused by the participants previous history of bone fractures and degree of skeletal deformities leading to loss of function and chronic pain, as indicated by others [20,21,22]. The severity of skeletal deformities is increased in the more severe phenotypes of the disease compared to the milder form [2]. A larger sample size would allow for better comparisons of the prevalence of MSK pain between groups. We must keep in mind that other studies have shown a poor correlation between clinical OI severity and the prevalence of MSK pain [6, 10, 11, 23].

The prevalence of MSK pain in our OI population during the past 7 days of 15–37% in the extremities and 37–56% in the back is higher than for a non-OI population. One study found an overall 14-day pain prevalence of 2.8–12.8% in the extremities (shoulder, elbow, hand/wrist, hip, knee, and foot/ankle) and 6.4–17.6% in the back (neck, lower, and upper back) in a Danish population of 4,817 healthy people aged 16 years of age or above [24]. When evaluating the prevalence of pain during the past 12 months among 20,173 Danish working people, the prevalence of MSK pain was higher, but not as high as in our OI Cohort. Upper-body (neck, shoulder, elbow, hand/wrist) and lower-body (lower back, hip, knee, and foot) MSK pain had a prevalence of 21.4% and 25.5% among working men with no mechanical exposure. For working women with no mechanical exposure, the 14-day MSK pain prevalence was 32.6% for upper-body pain and 28.9% for lower-body pain [25]. A comparative study including a representative and comparable reference population and a large cohort of adults with OI would allow us to evaluate how much higher the prevalence of MSK pain is in the OI population.

Our study found the highest MSK pain prevalence in the past 12 months in the lower back among participants with mild OI (93%) and a similarly high prevalence among participants with moderate to severe OI (85%), which corresponds with other studies of persons with OI [6, 7, 10, 13]. One study of a randomly chosen sample of the Dutch non-OI population found the lower back to be the most common site of MSK pain, with a prevalence of 21.2% [26]. A similar result was found in the Danish study including 20,173 working adults with a lower back MSK pain prevalence of 17.6% [24].

In our study, up to 48% of the participants with OI were missing out on regular work and daily activities due to site-specific MSK pain independent of clinical OI severity. This lack of correlation between clinical severity and the daily consequences of MSK pain and activities of daily living has been shown by others [10, 11]. However, physical Health-Related Quality of Life (HRQoL) has been shown to be significantly lower among adults with type III compared to those with OI type I and IV [11]. In this pilot study, we do not have the statistical power to evaluate between group differences in the consequences of MSK, but the differences in HRQoL observed by other may indicate that there an OI phenotype to MSK pain consequence correlation. Larger studies are needed to evaluate this.

This study had several limitations. First, the study population of 27 adults with OI is small, and our results may not represent the full Danish or larger international OI population. However, this was a pilot study to evaluate the feasibility of using the OI-NMQ in an OI population, and we have shown consistent results with the current literature by using this tool. Secondly, the OI-NMQ presumed the responder could recall their prevalence of MSK pain for the past 12 months, which might introduce recall bias. Former studies have, however, shown that the answers from the NMQ are reproducible regardless of recall bias [14, 19]. Third, selection bias might affect the results because the inclusion was done at the annual meeting of The Danish Osteogenesis Imperfecta Society. This may result in a study population with a different distribution of severity than what would be expected in the adult Danish OI population. Our study population may, therefore, not reflect the entire adult Danish OI population. However, our results are consistent with the current literature. Lastly, we chose to make missing answers indicative of a negative response, increasing the risk of underestimating the prevalence of MSK pain. Our results should, therefore, be seen as a conservative estimate of the prevalence and consequences of MSK pain in adults with OI. A larger sample would allow for subgroup analysis evaluating differences between clinical sub-types, age groups and genders.

The study is strengthened by the used of the OI-NMQ, a tool developed and adapted to evaluate the prevalence and consequences of MSK pain in an OI population. This is the first study to utilize the NMQ in evaluating MSK pain in adults with OI. Second, the questionnaire was short which made it an easy to understand and quick tool to complete among the participants. The response rate of 68% is acceptable since none of the participants at The Danish Osteogenesis Imperfecta Society annual meeting was pre-informed of the OI-NMQ, and the questionnaire had to be answered and returned by the end of the meeting. At last, the study is strengthened by the close collaboration with The Danish Osteogenesis Imperfecta Society, who have been instrumental in developing the questionnaire, data collection, interpretation, and dissemination.