Abstract
Context:
The OSTEOPATHIC Trial demonstrated substantial improvement in pain intensity, decreased need for rescue medication for pain, and greater likelihood of recovery in patients with chronic low back pain who received 6 osteopathic manipulative treatment (OMT) sessions over 3 months.
Objective:
To assess osteopathic medical care and the effectiveness of OMT for chronic low back pain in a real-world setting without the constraints of a rigid research protocol.
Methods:
An observational study of 445 adults with chronic low back pain who had an established osteopathic physician (ie, DO) or allopathic physician (ie, MD) was conducted within the PRECISION Pain Research Registry from April 2016 through February 2019. Primary outcome measures included a numerical rating scale for low back pain intensity, the Roland-Morris Disability Questionnaire for back-related functioning, and use of nonsteroidal anti-inflammatory drugs or opioids for low back pain.
Results:
A total of 79, 48, and 318 patients, respectively, were treated by DOs who used OMT, DOs who did not use OMT, or MDs. Patients treated by DOs who used OMT reported significantly lesser low back pain intensity (mean numerical rating score, 5.6; 95% CI, 5.1-6.1 vs 6.1; 95% CI, 5.9-6.3; P=.04) and back-related disability (mean Roland-Morris Disability score, 12.4; 95% CI, 11.1-13.8 vs 14.4; 95% CI, 13.7-15.0; P=.009) than patients treated by MDs. Patients treated by DOs who used OMT also reported less frequent use of nonsteroidal anti-inflammatory drugs (multivariate odds ratio, 0.41; 95% CI, 0.24-0.70; P=.001) or opioids (multivariate odds ratio, 0.52; 95% CI, 0.28-0.98; P=.04). There were no significant differences in primary outcomes between DOs who did not use OMT and MDs.
Conclusion:
This study of community-based patients in a pain research registry supports the effectiveness of OMT as an integral component of osteopathic medical care for chronic low back pain. Patients treated by DOs who did not use OMT did not experience better results than patients treated by MDs in any primary outcome measure. Further research is needed to more specifically compare the effects of OMT with other treatment effects that may be attributed to patient-DO interactions during medical encounters.
Randomized controlled trials,
1-3 systematic reviews and meta-analyses,
4,5 and national health care surveys
6,7 in the United States provide varying levels of support for the view that osteopathic medical care, including osteopathic manipulative treatment (OMT), is a particularly effective intervention for low back pain. Based largely on findings from the first systematic review and meta-analysis conducted to assess this research question,
4 the only clinical practice guideline developed by the American Osteopathic Association recommends using OMT in patients wherein somatic dysfunction is a cause of, or contributing factor to, low back pain.
8
Specifically, with regard to chronic low back pain, the OSTEOPATHIC Trial has provided the most compelling evidence to date that supports the efficacy of OMT.
3,9 The OSTEOPATHIC Trial demonstrated substantial improvement in pain intensity, decreased need for rescue medication for pain, and greater likelihood of recovery in patients with chronic low back pain who received 6 OMT sessions during a 3-month period.
3,9 Nevertheless, the extent to which such findings are generalizable to other patients with chronic low back pain remains unclear. The National Ambulatory Medical Care Survey (NAMCS) has been used in efforts to address this question. Therein, from 2003 to 2004 it was shown that osteopathic physicians (ie, DOs) provided a disproportionately large proportion of medical care visits for chronic low back pain.
7 Osteopathic physicians also prescribed pharmacologic therapy less often and maintained greater continuity of care than allopathic physicians (ie, MDs) in another NAMCS study
6 of patients with low back pain from 2002 to 2006. However, the latter findings were not specific to chronic low back pain. Additionally, osteopathic research using NAMCS is limited in that it does not provide specific data on the use of OMT within patient visits. The purpose of the present study was to address such limitations and voids in the current osteopathic literature pertaining to the use of OMT for chronic low back pain in real-world settings by using data collected from a community-based pain research registry.
Patients completed the data collection instruments using a computer or mobile device provided by project personnel and the Qualtrics Survey Software. Research coordinators assisted patients if they had difficulty reading or understanding the questions. The electronic data collection system permitted patients to take breaks while answering questions if needed; however, it required completion of all items on the research instruments. Qualtrics survey data were exported to the SPSS version 23 software (IBM), which was subsequently used for data management and analysis. Descriptive statistics were computed using numbers and percentages for nominal and dichotomous variables and mean (SD) for continuous variables. Statistical comparisons of patients treated by physicians within each of the 3 categories (DOs who used OMT, DOs who did not use OMT, or MDs) were performed using contingency table methods for categorical variables or analysis of variance with the Fisher least significant difference test for post hoc comparisons for continuous variables. Additionally, analysis of covariance was performed to adjust the low back pain intensity and back-related functioning outcomes for patient age and sex. Simple logistic regression models were initially used to compute odds ratios (ORs) and 95% CIs for patient-reported use of NSAIDs or opioids according to the type of treating physician, with MDs used as the reference category. Multiple logistic regression was subsequently used to adjust for potential confounders. Therein, the age, the NRS score for low back pain intensity, and the Roland-Morris Disability score were entered as covariates, and the ORs and 95% CIs for use of the relevant drug category (NSAIDs or opioids) were adjusted for current use of the alternative drug category (opioids or NSAIDs). Hypotheses were tested at the .05 level of statistical significance.
Nonsteroidal anti-inflammatory drugs were reportedly used for low back pain by 41 (51.9%), 30 (62.5%), and 217 (68.2%) patients treated by DOs who used OMT, DOs who did not use OMT, or MDs, respectively. The corresponding summary measures for NSAID use in comparison with patients treated by MDs were as follows: OR, 0.50; 95% CI, 0.30-0.83;
P=.007 for patients treated by DOs who used OMT; and OR, 0.78; 95% CI, 0.41-1.46;
P=.43 for patients treated by DOs who did not use OMT. Patients treated by DOs who used OMT also less frequently reported using NSAIDs in the multivariate analysis that controlled for potential confounders (OR, 0.41; 95% CI, 0.24-0.70;
P=.001) (
Table 2). Increasing age, being a male patient, having a diagnosis of a herniated disc, and currently using opioids were other patient factors significantly associated with reporting less frequent use of NSAIDs for low back pain.
Table 2.
Multiple Logistic Regression for Factors Associated With Current Use of Nonsteroidal Anti-Inflammatory Drugs or Opioids for Chronic Low Back Pain (N=445)a
| Characteristic | Current User |
| NSAIDs | Opioids |
| OR (95% CI) | P Value | OR (95% CI) | P Value |
| Type of Physician | | | | |
| MD | 1.00 | … | 1.00 | … |
| DO who used OMT | 0.41 (0.24-0.70) | .001 | 0.52 (0.28-0.98) | .04 |
| DO who did not use OMT | 0.70 (0.36-1.36) | .30 | 0.79 (0.38-1.63) | .52 |
| Age | 0.98 (0.96-1.00) | .047 | 1.01 (0.99-1.03) | .24 |
| Sex | | | | |
| Female | 1.00 | … | 1.00 | … |
| Male | 0.62 (0.40-0.97) | .04 | 0.90 (0.55-1.48) | .69 |
| Cigarette Smoking Status | | | | |
| Never or Former Smoker | 1.00 | … | 1.00 | … |
| Current Smoker | 0.91 (0.52-1.57) | .72 | 2.13 (1.21-3.75) | .009 |
| Presence of Widespread Pain | | | | |
| No | 1.00 | … | 1.00 | … |
| Yes | 0.87 (0.53-1.44) | .60 | 1.81 (1.09-3.00) | .02 |
| Diagnosis of a Herniated Disc | | | | |
| No | 1.00 | … | 1.00 | … |
| Yes | 0.55 (0.35-0.87) | .01 | 2.45 (1.54-3.91) | <.001 |
| Diagnosis of Sciatica | | | | |
| No | 1.00 | … | 1.00 | … |
| Yes | 1.43 (0.91-2.24) | .12 | 1.41 (0.88-2.24) | .15 |
| Diagnosis of Depression | | | | |
| No | 1.00 | … | 1.00 | … |
| Yes | 1.18 (0.76-1.83) | .46 | 1.20 (0.76-1.90) | .44 |
| Low Back Pain Intensity | 0.97 (0.86-1.08) | .56 | 0.95 (0.84-1.08) | .41 |
| Back-Related Disability | 1.01 (0.97-1.06) | .56 | 1.10 (1.05-1.15) | <.001 |
| Currently Uses NSAID | | | | |
| No | … | … | 1.00 | … |
| Yes | … | … | 0.55 (0.34-0.87) | .01 |
| Currently Uses Opioids | | | | |
| No | 1.00 | … | … | … |
| Yes | 0.55 (0.35-0.87) | .01 | … | … |
Table 2.
Multiple Logistic Regression for Factors Associated With Current Use of Nonsteroidal Anti-Inflammatory Drugs or Opioids for Chronic Low Back Pain (N=445)a
| Characteristic | Current User |
| NSAIDs | Opioids |
| OR (95% CI) | P Value | OR (95% CI) | P Value |
| Type of Physician | | | | |
| MD | 1.00 | … | 1.00 | … |
| DO who used OMT | 0.41 (0.24-0.70) | .001 | 0.52 (0.28-0.98) | .04 |
| DO who did not use OMT | 0.70 (0.36-1.36) | .30 | 0.79 (0.38-1.63) | .52 |
| Age | 0.98 (0.96-1.00) | .047 | 1.01 (0.99-1.03) | .24 |
| Sex | | | | |
| Female | 1.00 | … | 1.00 | … |
| Male | 0.62 (0.40-0.97) | .04 | 0.90 (0.55-1.48) | .69 |
| Cigarette Smoking Status | | | | |
| Never or Former Smoker | 1.00 | … | 1.00 | … |
| Current Smoker | 0.91 (0.52-1.57) | .72 | 2.13 (1.21-3.75) | .009 |
| Presence of Widespread Pain | | | | |
| No | 1.00 | … | 1.00 | … |
| Yes | 0.87 (0.53-1.44) | .60 | 1.81 (1.09-3.00) | .02 |
| Diagnosis of a Herniated Disc | | | | |
| No | 1.00 | … | 1.00 | … |
| Yes | 0.55 (0.35-0.87) | .01 | 2.45 (1.54-3.91) | <.001 |
| Diagnosis of Sciatica | | | | |
| No | 1.00 | … | 1.00 | … |
| Yes | 1.43 (0.91-2.24) | .12 | 1.41 (0.88-2.24) | .15 |
| Diagnosis of Depression | | | | |
| No | 1.00 | … | 1.00 | … |
| Yes | 1.18 (0.76-1.83) | .46 | 1.20 (0.76-1.90) | .44 |
| Low Back Pain Intensity | 0.97 (0.86-1.08) | .56 | 0.95 (0.84-1.08) | .41 |
| Back-Related Disability | 1.01 (0.97-1.06) | .56 | 1.10 (1.05-1.15) | <.001 |
| Currently Uses NSAID | | | | |
| No | … | … | 1.00 | … |
| Yes | … | … | 0.55 (0.34-0.87) | .01 |
| Currently Uses Opioids | | | | |
| No | 1.00 | … | … | … |
| Yes | 0.55 (0.35-0.87) | .01 | … | … |
×
Opioids were reportedly used for low back pain by 20 (25.3%), 17 (35.4%), and 127 (39.9%) patients treated by DOs who used OMT, DOs who did not use OMT, or MDs, respectively. The corresponding summary measures for opioid use in comparison with patients treated by MDs were as follows: OR, 0.51; 95% CI, 0.29-0.89; P=.02 for patients treated by DOs who used OMT; and OR, 0.82; 95% CI, 0.44-1.55 P=.55 for patients treated by DOs who did not use OMT. Patients treated by DOs who used OMT also less frequently reported using opioids in the multivariate analysis (OR, 0.52; 95% CI, 0.28-0.98; P=.04). Current cigarette smoking, presence of widespread pain, having a diagnosis of a herniated disc, and increasing levels of back-related disability were patient factors significantly associated with reporting more frequent use of opioids for low back pain, whereas reported current use of NSAIDs was associated with less frequent use of opioids.