Symptomatic Reactions, Clinical Outcomes and Patient Satisfaction Associated with Upper Cervical Chiropractic Care: A Prospective, Multicenter, Cohort Study

The Chiro.Org Blog

SOURCE:   BMC Musculoskelet Disord. 2011 (Oct 5); 12: 219

Kirk Eriksen, Roderic P Rochester, and Eric L Hurwitz

Chiropractic Health Institute, PC, Clinic Director, 2500 Flowers Chapel Road, Dothan, AL 36305, USA.

BACKGROUND:   Observational studies have previously shown that adverse events following manipulation to the neck and/or back are relatively common, although these reactions tend to be mild in intensity and self-limiting. However, no prospective study has examined the incidence of adverse reactions following spinal adjustments using upper cervical techniques, and the impact of this care on clinical outcomes.

METHODS:   Consecutive new patients from the offices of 83 chiropractors were recruited for this practice-based study. Clinical outcome measures included:

  1. Neck pain disability index (100-point scale),
  2. Oswestry back pain index (100-point scale),
  3. 11-point numerical rating scale (NRS) for neck, headache, midback, and low back pain,
  4. treatment satisfaction, and
  5. Symptomatic Reactions (SR).

Data were collected at baseline, and after approximately 2 weeks of care. A patient reaching sub-clinical status for pain and disability was defined as a follow-up score <3 NRS and <10%, respectively. A SR is defined as a new complaint not present at baseline or a worsening of the presenting complaint by >30% based on an 11-point numeric rating scale occurring <24 hours after any upper cervical procedure. RESULTS:   A total of 1,090 patients completed the study having 4,920 (4.5 per patient) office visits requiring 2,653 (2.4 per patient) upper cervical adjustments over 17 days. Three hundred thirty- eight (31.0%) patients had SRs meeting the accepted definition. Intense SR (NRS ≥8) occurred in 56 patients (5.1%). Outcome assessments were significantly improved for neck pain and disability, headache, mid-back pain, as well as lower back pain and disability (p <0.001) following care with a high level (mean = 9.1/10) of patient satisfaction. The 83 chiropractors administered >5 million career upper cervical adjustments without a reported incidence of serious adverse event.

CONCLUSIONS:   Upper cervical chiropractic care may have a fairly common occurrence of mild intensity SRs short in duration (<24 hours), and rarely severe in intensity; however, outcome assessments were significantly improved with less than 3 weeks of care with a high level of patient satisfaction. Although our findings need to be confirmed in subsequent randomized studies for definitive risk-benefit assessment, the preliminary data shows that the benefits of upper cervical chiropractic care may outweigh the potential risks.

From the FULL TEXT Article:


This study provides the incidence of Symptomatic Reactions (SRs) related to upper cervical techniques (UCT) and the improvement in clinical outcomes following this form of spinal care. The preponderance of the evidence shows that major complications resulting from chiropractic care are very rare, although transient discomfort and other minor side effects of chiropractic care are common. Various prospective studies have shown about 30% to 60% of patients receiving SMT experience minor side effects shortly after treatment [1-12,38]. Local discomfort in the treatment area accounts for half to two-thirds of the reactions. Other effects such as radiating symptoms, headaches, tiredness/fatigue each account for about 10% of SRs. Side effects such as nausea, dizziness and other symptoms are rare and most studies show that each of these comprise of 5% or less of reported symptoms. The majority of reactions have been reported to begin within 24 hours of the treatment visit and to resolve in less than 24 hours.

Table 1.   Doctor and patient frequencies, by type of UCT

Table 2.   Frequency distributions and/or means (SDs) of selected sociodemographic, clinical, and health-care variables

Table 3.   Frequency distributions of presenting chief complaints

Table 4.   Frequency distributions and/or means (SDs) of clinical outcome variables at baseline and following an average of 17 days, 4.5 visits and 2.4 upper cervical adjustments

Table 5.   Frequency distributions and means (SDs) of symptomatic reactions (SRs), by type of reaction

Table 6.   Estimated effects (risk ratios [RRs] and 95% confidence intervals [CIs]) of SRs on levels of satisfaction* (n = 1089)

Our current study compares favorably to most of these results. Thirty-one percent of the patients had SRs that were defined as the following:

  • A worsening of presenting chief complaint of >30%
  • A SR that has discomfort rating of >1
  • A SR that began ≤24 hours from the patient’s adjustment

This study compared the musculoskeletal SRs with the patients’ presenting complaints. By using the clinical data of patients that presented with musculoskeletal symptoms (neck, headache, midback and/or lowback pain) (1035/1090 which is 95% of the patients) it was determined that only 12.3% of the patients had NRS that were >30% compared to their initial discomfort ratings (using the 11-point NRS for pain). While analyzing clinical data measurements this number may represent (1) SR, (2) the care failure rate or (3) a combination of the two. The study found the incidence of intense SR (≥8 on the NRS) was only 6.6% of the total number of SRs and occurred in 5.1% of patients. It should be noted the 83 chiropractors in the study were asked to estimate the number of upper cervical adjustments they had performed in their career. The total estimation for career adjustments came to 5,085,014. The doctors did not report any serious AE ever occurring in their practices (i.e., strokes or permanent injuries).

It should be noted that several of the side effects reported in our study have been found in previous studies with patients taking medication. Headaches, fatigue, dizziness, and nausea are among the most common drug-related adverse reactions [39], and these have been reported by people not taking any medication [40,41] The tenuous association between cervical SMT and cerebrovascular incidents (CVI) has been reviewed by various studies, and appears to have an estimated occurrence ranging from no causative association, to 1 in 300,000 to 500,000 [42,43], to 1.3 million [44], or 5.85 million cervical manipulations [45]. There appears to be a link between upper cervical rotation SMT and cerebrovascular incidents [46,47]; however, two cases have been reported in the literature involving strokes following non-force and neutral position cervical manipulative procedures [48,49]. Cassidy et al. [50] found that the risk of having a stroke was equal between patients consulting a chiropractor or general medical practitioner. This suggests that cervical SMT may not be a cause of cerebrovascular accidents, but associated with a stroke in progress. Our present study was not designed specifically to detect rare major complications, although doctors were responsible for reporting any and all AEs that occurred during the course of care.

The current study’s outcome statistics are very similar to the results of a smaller case series (N = 66) studying neck pain outcomes following a low-force UCT [34]. The case series showed 34.8% of cases achieved a normal NDI (<5 or <10 for the 50 or 100 point scale, respectively) after 13.6 days, 5.7 office visits and 2.7 UC adjustments from a baseline of 3%, a 31.4% net gain. The current study shows 42.4% at a normal NDI from a baseline of 10.3% a 32% net gain. One prior study with a large number of patients (N = 529) studying AEs and outcomes following SMT demonstrated an 11.6% net gain score of NDI in the short term (4th visit) and 31.4% net gain in the long term (12 months/9.3 SMT treatments) [5].

The International Chiropractors Association published ‘Best Practices and Practice Guidelines’ in which they summarized multiple randomized controlled trials that used SMT or mobilization as the primary form of care and NRS or visual analog scale (VAS) as outcomes [51]. They summarize 66 RCTs for low back pain including a total of 4,661 patients, with 36 RCTs reporting NRS or VAS. Twenty-four RCTs involved chiropractors. The average number of office visits was 8.3 reporting a 42.6% improved NRS for pain on average. For neck, upper back and headache pain they summarize 54 RCTs (23 with chiropractors) including 2,069 patients reporting an average improvement of 46.5% with 7.7 office visits [5,8,52-55]. The current study differs in that it is not an RCT but a prospective, multicenter observational cohort study and cannot establish a causative association; however, it is similar in nature due to the large numbers of patients involved. A direct comparison cannot be made due to differing patient populations and other reasons.

It can be said that UCT fairs well when judged against the published guidelines and other studies in terms of patient safety and clinical efficacy. One of the chiropractic premises is that the correction or reduction of relative vertebral misalignment(s) in the upper cervical spine is clinically important. However, some believe the concepts are invalid, insignificant or implausible as it relates to outcomes, especially for spinal conditions like lower back pain because it is not adjacent to the upper cervical region. The data in the current study demonstrates a 57% average improvement for lower back pain following UCT but reports only 1.4% SRs in the lower back region. This data stimulates the question: “Is UCT a plausible type of care for lower back pain patients?” Two case series and one randomized clinical trial have demonstrated better outcomes for musculoskeletal symptoms and hypertension (without pain), respectively, for patients receiving reductions in C1 misalignment following UCT care relative to patients receiving smaller or no corrections in alignment [33,34,56]. Also, the present study reports fewer adjustments and a shorter follow-up period than previous chiropractic studies while demonstrating similar or better-improved outcome levels [52-55].

Two studies have demonstrated a statistically better clinical outcome in patients following Grostic/Orthospinology care [34,56] . This was achieved in the patients with a significant reduction in upper cervical misalignments as measured on their initial post adjustment radiographs. The patients with higher percent corrections on the first visit had a decreased need for follow-up visits and upper cervical adjustments. The subjects receiving higher upper cervical misalignment corrections also demonstrated higher improvement in various clinical outcomes compared to those having a lesser degree of correction. However, both of these studies were retrospective in design and did not have control groups.

Patient satisfaction in this study was very high with an overall rating of 9.1 on the 11-point NRS. Hertzman-Miller, et al. [57] found an average patient satisfaction rating of 30.6 (7.1)/36.1 (5.4) for medical and chiropractic care, respectively, for lower back pain patients on a scale of 10 (least satisfied) to 50 (most satisfied). Rubinstein et al. [5] reports a patient satisfaction rating of 7.8 (1.8) on an 11-point NRS for neck pain patients undergoing a trial of primarily diversified chiropractic care after 3 visits. Hurwitz et al. [58] found greater satisfaction increased the odds of remission from clinically meaningful pain and disability at 6 weeks but not at 6, 12 or 18 months. The current study revealed higher risk ratios for a satisfaction rating of poor for patients with any SR and intense SRs after an average of 17 days of care. Other studies and our data indicate that SRs and outcomes for pain and dysfunction are inversely associated in the short term while patient satisfaction and outcomes have a direct relationship.

This study has several strengths, including a large sample size, wide geographic representation, high participation and follow-up rates, and use of validated outcome measures; however, the observational (non-randomized) nature of the study precludes firm causal inferences of upper cervical chiropractic care with clinical outcomes, symptomatic reactions and patient satisfaction. Because the study lacks a control group, we could not estimate risks and rates of outcomes relative to other types of care or to a no-treatment condition. Also, our findings may not necessarily apply to the provision of other types of chiropractic care, care delivered in other settings, or in populations of patients with other conditions, comorbidities, or with different prognoses. Nevertheless, our study offers a valid description of the symptomatic reactions, clinical outcomes and patient satisfaction associated with UC chiropractic care in a cohort of over 1,000 patients.


Chiropractic care using UCT may have a fairly common occurrence of mild intensity SRs short in duration (<24 hours), and rarely severe in intensity; however, outcome assessments were significantly improved with less than 3 weeks of care with a high level of patient satisfaction. Although our findings need to be confirmed in subsequent randomized studies for definitive risk-benefit assessment, the preliminary data shows that the benefits of UC chiropractic care may outweigh the potential risks.