The epidemiology of osteopathic diagnoses and treatments in United States emergency departments from 2018 to 2021

Methods

Results

We identified 9,576 ED encounters with OMM in the study period. Among all EDs, the weighted frequency of encounters with OMM per 100,000 ED encounters was 6.9 (95 % CI, 3.8 to 11.5) for 2018, 8.6 (4.3–15.5) for 2019, 12.6 (3.5–32.0) for 2020, and 5.5 (3.3–8.5) for 2021 (Table 1). The point estimates for the weighted frequencies were numerically greater for both the musculoskeletal pain and lower back pain subgroups among all EDs and both ED subsets (Figure 1, Table 1, and Appendix 2–3 in Supplementary Material).

Figure 1:

The plot displays the weighted frequencies of encounters with OMM per 100,000 ED encounters, faceted by ED group. ED groups included all EDs, those with at least one encounter with OMM in the given year, and those with counts of encounters with OMM in the given year greater than the upper quartile plus 1.5 times the IQR. The low back pain and musculoskeletal pain subgroups include encounters with at least one ICD-10-CM code within the CCSR codes for low back pain (MUS038) or musculoskeletal pain, no low back pain (MUS010). ED, emergency department; CCSR, clinical classifications software refined; CI, confidence interval; IQR, interquartile range; OMM, osteopathic manipulative medicine.

Encounters with OMM were more common among the ED subsets than among all EDs (Figure 1, Table 1, and Appendix 2–3 in Supplementary Material). Among EDs with OMM encounters, the weighted median number of encounters with OMM were 1 [IQR 1 to 3] in 2018, 2 [1, 4] in 2019, 2 [1, 4] in 2020, and 2 [1, 3] in 2021. Among EDs with frequent OMM encounters, the median number of encounters with OMM were 18 [IQR 12, 42] in 2018, 21 [11, 41] in 2019, 19 [11, 53] in 2020, and 16 [10, 22] in 2021 (Table 2). A total of 7,565 of 9,576 (79.0 %) encounters with OMM were from this subgroup of EDs with frequent OMM encounters.

Most encounters were for female patients 5,669 (59.2 %), and the median age of patients from the identified encounters with OMM was 53 (IQR 36 to 67). The most common body system for the primary diagnosis was diseases of the musculoskeletal system and connective tissue with 2,349 (24.5 %) encounters. Just under half of the encounters with OMM were admitted and occurred in the Midwest US. The majority were from metropolitan teaching hospitals (7,094, 74.1 %), and most only had osteopathic ICD-10-CM diagnostic codes (8,064, 84.2 %) (Table 3). The most common osteopathic ICD-10-CM diagnosis code was for segmental and somatic dysfunction of the thoracic region, with 4,202 of 21,358 (19.7 %) codes (Appendix 4 in Supplementary Material). The majority of CPT/HCPCS codes were for osteopathic manipulative treatment involving 1–2 body regions (652 of 838 [77.8 %] codes) (Appendix 5 in Supplementary Material). We observed a wide variety of ICD-10-PCS codes (Appendix 6 in Supplementary Material).

Discussion

Encounters with OMM were rare in US EDs, occurring in up to 12.6 ED encounters per 100,000 (95 % CI, 3.5 to 32.0) (Figure 1, Table 1). However, they were more common among select EDs (i.e., EDs with frequent OMM encounters), occurring in up to 280.9 ED encounters per 100,000 (95 % CI, 2.8 to 1866.3) (Figure 1, Appendix 3 in Supplementary Material). The frequency was higher among encounters with at least one musculoskeletal pain or low back pain diagnosis (Figure 1, Table 1, and Appendix 2–3 in Supplementary Material). For perspective, endotracheal intubation, a core skill learned in emergency medicine residency [30], occurs in 270 ED encounters per 100,000 [31]. Furthermore, the median count of encounters with OMM was only 1 to 2 per ED per year among EDs with OMM encounters, and most encounters occurred from the small group of EDs who frequently performed OMM (Table 2).

The very low frequency of osteopathic diagnoses and treatments among US ED patients is likely multifactorial. First, the study period coincides with unprecedented changes in ED operations due to ED crowding and the coronavirus disease 2019 (COVID-19) pandemic [32], 33]. ED crowding can significantly compromise the quality of ED care. For example, ED crowding has been associated with reduced compliance with severe sepsis and septic shock care [34], which are diagnoses with a high risk of mortality [35]. Therefore, interventions and care, like OMM, which are not expected to impact short-term mortality or significant morbidity, may not have been prioritized. Although unknown, OMM may be more frequently utilized in the urgent care setting with lower acuity patients or in non-hospital-owned EDs (e.g., Veterans Affairs hospitals) with more resources relative to patient acuity. Furthermore, the risk of transmission of COVID-19 may have discouraged the use of OMM during the later study period, even among patients without typical symptoms given the frequency of incidental infections [36]. However, there did not appear to be a decline in the frequency of encounters with OMM in 2020 or 2021 compared to 2019 (Figure 1, Table 1, and Appendix 2–3 in Supplementary Material). Second, advanced practice providers (APPs) care for many ED patients, especially those with low-acuity complaints [37]. These patients with lower-risk complaints cared for by APPs may be better candidates for OMM but are not triaged to emergency physician team members (i.e., DOs), who are generally responsible for the more critically ill ED patients. APPs may be in a better position to utilize OMM in the ED if they can obtain the necessary training and supervision. Further work is needed to identify how OMM can be incorporated into unscheduled acute care in the ED and other acute care settings.

We observed some interesting findings that inform where osteopathic diagnoses and treatments are most utilized. First, 47.7 % of encounters with OMM were admitted (Table 3). For these admitted cases, it is ambiguous whether the osteopathic diagnosis(es) or treatment(s) occurred in the ED or inpatient setting because ICD-10 codes apply to the entire hospital stay for admitted patients in this dataset. Therefore, the frequency of ED encounters with OMM may be lower than measured if inpatient clinicians performed the osteopathic diagnoses and treatments for these patients. Second, 74.1 % of encounters with OMM were from EDs at metropolitan teaching hospitals (Table 3). Therefore, teaching environments may facilitate the use of OMM more often than nonteaching settings. Lastly, most encounters had osteopathic diagnostic codes (ICD-10-CM) alone (84.2 %), suggesting that OMM is utilized primarily as a diagnostic intervention among ED patients (Table 3). Future work may aim to clarify where OMM occurs more frequently for patients admitted from the ED (i.e., in the ED or inpatient unit) and whether attendings alone or attendings with trainees chose to utilize OMM more often.

Lastly, these results may have some nonclinical implications. Given the infrequency of OMM use in the ED, DO emergency physicians may benefit less from research grants provided by osteopathic institutions that specifically fund OMM research rather than DO-lead research. Also, emergency medicine residency programs may find less benefit from osteopathic recognition than other specialties where the clinical environment better facilitates the use of OMM. In addition, the majority of US states have a combined medical board that licenses both MDs and DOs, whereas a minority have separate boards [18], 38]. These results provide additional evidence that OMM is performed infrequently [15], which may inform decisions by states with separate licensing boards considering a transition to a single board.

Conclusions

OMM is infrequently provided among encounters from hospital-owned EDs in the US. Further research is needed to describe the frequency of OMM across all acute care settings and to investigate opportunities to incorporate OMM into the ED workflow.