Advertisement

Identifying Patient Characteristics Associated With the Occurrence of Post Treatment Non-serious Adverse Events After Cervical Spine Manual Therapy Treatment in Patients With Neck Pain

Open AccessPublished:August 26, 2022DOI:https://doi.org/10.1016/j.apmr.2022.08.007

      Abstract

      Objectives

      To compare prevalence rates of serious and non-serious adverse events after manipulation and mobilization and to identify risk factors of serious and non-serious adverse events following 4 types of manual therapy treatment in patients with neck pain.

      Design

      A prospective cohort study in primary care manual therapy practice.

      Participants

      Patients with neck pain (N=686) provided data on adverse events after 1014 manipulation treatments, 829 mobilization treatments, 437 combined manipulation and mobilization treatments, and 891 treatments consisting of “other treatment modality”.

      Interventions

      Usual care manual therapy.

      Main Outcome Measures

      A chi-square test was performed to explore differences in prevalence rates. Logistic regression analysis was performed within the 4 treatment groups. A priori we defined associations between patient-characteristics and adverse events of odds ratio (OR)>2 or OR<0.5 as clinically relevant.

      Results

      No serious adverse events, such as cervical artery dissection or stroke, were reported. With regard to non-serious adverse events, we found that these are common after manual therapy treatment: prevalence rates are ranging from 0.3% to 64.7%. We found a statistically significant difference between the 4 types of treatments, detrimental to mobilization treatment. Logistic regression analysis resulted in 3 main predictors related to non-serious adverse events after manual therapy treatment: smoking (OR ranges from 2.10 [95% confidence interval [CI] 1.37-3.11] to 3.33 [95% CI 1.83-5.93]), the presence of comorbidity (OR ranges from 2.32 [95% CI 1.22-4.44] to 3.88 [95% CI 1.62-9.26]), and female sex (OR ranges from 0.22 [95% CI 0.11-0.46] to 0.49 [95% CI 0.28-0.86]).

      Conclusion

      There is a significant difference in the occurrence of non-serious adverse events after mobilization compared with manipulation or a combination of manipulation and mobilization. Non-serious adverse events in manual therapy practice are common and are associated with smoking and the presence of comorbidity. In addition, women are more likely to report non-serious adverse events.

      Keywords

      List of abbreviations:

      AE (adverse event), CI (confidence interval), FABQ (Fear Avoidance Beliefs Questionnaire), GPE (global perceived effect), HVT (high velocity thrust), MOB (mobilization), MP (manipulation), MT (manual therapy), NBQ (Neck Bournemouth Questionnaire), NDI (Neck Disability Index), NRS (Numeric Rating Scale for pain), OR (odds ratio)
      Cervical spine manipulation and cervical spine mobilization are commonly used (treatment) modalities for the treatment of patients with neck pain. An Australian survey among manipulative therapists shows that cervical spine manipulative techniques are used by 84.5% of manipulative therapists in patients with neck pain.
      • Magarey ME
      • Rebbeck T
      • Coughlan B
      • Grimmer K
      • Rivett DA
      • Refshauge K.
      Pre-manipulative testing of the cervical spine review, revision and new clinical guidelines.
      Six to 12% of the population with neck pain undergoes cervical spine manipulation annually.
      • Hurwitz EL.
      Epidemiology: spinal manipulation utilization.
      Manual therapy (MT) is beneficial to patients with neck pain,
      • Gross AR
      • Miller J
      • D'Sylva J
      • et al.
      Manipulation or mobilisation for neck pain: a Cochrane review systematic review.
      ,
      • Bronfort G
      • Evans R
      • Anderson AV
      • Svendsen KH
      • Bracha Y
      • Grimm RH.
      Spinal manipulation, medication, or home exercise with advice for acute and subacute neck pain: randomized trial.
      ,
      • Chaibi A
      • Russell MB.
      Manual therapies for cervicogenic headache.
      especially when used in combination with exercise.
      • Chaibi A
      • Russell MB.
      Manual therapies for cervicogenic headache.
      MT in The Netherlands may consist of high velocity thrust (HVT) manipulation or low velocity non-thrust mobilization, but also exercise to stabilize the spine and other modalities. Despite evidence to support the benefits of cervical HVT manipulation, the use of this treatment modality still remains controversial because of its potential risks of serious adverse events (AEs).
      • Cagnie B
      • Vinck E
      • Beernaert A
      • Cambier D.
      How common are side effects of spinal manipulation and can these side effects be predicted?.
      ,
      • Ernst E.
      Vascular accidents after neck manipulation: cause or coincidence?.
      ,
      • Rubinstein SM
      • Leboeuf-Yde C
      • Knol DL
      • De Koekkoek TE
      • Pfeifle CE
      • Van Tulder ME.
      The benefits outweigh the risks for patients undergoing chiropractic care for neck pain: a prospective, multicenter, cohort study.
      AEs after interventions to the cervical spine are commonly divided into 2 groups: serious and non-serious. There is no clear consensus concerning the definition of both serious
      • Malone DG
      • Baldwin NG
      • Tomecek FJ
      • et al.
      Complications of cervical spine manipulation therapy: 5-year retrospective study in a single-group practice.
      ,
      • Thiel HW
      • Bolton JE
      • Docherty S
      • et al.
      Safety of chiropractic manipulation of the cervical spine: a prospective national survey.
      ,
      • Dvorak J
      • Orelli FV.
      How dangerous is manipulation of the cervical spine?.
      and non- serious AE.
      • Cagnie B
      • Vinck E
      • Beernaert A
      • Cambier D.
      How common are side effects of spinal manipulation and can these side effects be predicted?.
      ,
      • Carnes D
      • Mullinger B
      • Underwood M.
      Defining adverse events in manual therapies: a modified Delphi consensus study.
      ,
      • Carlesso LC
      • Gross AR
      • Santaguida PL
      • Burnie S
      • Voth S
      • Sadi J.
      Adverse events associated with the use of cervical manipulation and mobilization for the treatment of neck pain in adults: a systematic review.
      But, commonly, serious AE refer to events resulting in life-threatening situations or even in death, while the non-serious AE are benign in nature and of short duration (resolve within 24 h) non-serious AEs can then be further divided into common and uncommon reactions based upon the frequency of their occurrence.
      • Cagnie B
      • Vinck E
      • Beernaert A
      • Cambier D.
      How common are side effects of spinal manipulation and can these side effects be predicted?.
      ,
      • Senstad O
      • Leboeuf-Yde C
      • Borchgrevnink C.
      Predictors of side effects to spinal manipulative therapy.
      Fortunately, the occurrence of serious AE (such as cervical artery dissection, stroke, or transient ischemic attack) after cervical spinal manipulation is rare.
      • Carlesso LC
      • Gross AR
      • Santaguida PL
      • Burnie S
      • Voth S
      • Sadi J.
      Adverse events associated with the use of cervical manipulation and mobilization for the treatment of neck pain in adults: a systematic review.
      ,
      • Boyle E
      • Côté P
      • Grier AR
      • Cassidy JD.
      Examining vertebrobasilar artery stroke in two Canadian provinces.
      Incidence rates range from 1 in 100,000 to 1 in several million.
      • Cassidy JD
      • Boyle E
      • Côté P
      • et al.
      Risk of vertebral stroke and chiropractic care: results of a population-based case control and case cross over study.
      There is, however, evidence to suggest that the reporting of AE is too poor and incomplete to enable any meaningful interpretation of risk.
      • Carlesso LC
      • Gross AR
      • Santaguida PL
      • Burnie S
      • Voth S
      • Sadi J.
      Adverse events associated with the use of cervical manipulation and mobilization for the treatment of neck pain in adults: a systematic review.
      ,
      • Ernst E
      • Canter PH.
      A systematic review of systematic reviews of spinal manipulation.
      ,
      • Ernst E
      • Posadzki P.
      Reporting of adverse effects in randomised clinical trials of chiropractic manipulations: a systematic review.
      ,
      • Wynd S
      • Estaway M
      • Vohra S
      • Kawchuk G.
      Correction: the quality of reports on cervical arterial dissection following cervical spinal manipulation.
      Serious AEs were considered to be primarily—and often even exclusively—related to manipulation techniques.
      • Kerry R
      • Taylor A.
      Cervical spine pre-treatment screening for arterial dysfunction: out with the old, in with the new.
      Reported risk factors for serious AE include age, biological sex, hypertension, diabetes, migraines, use of oral contraceptives, and smoking.
      • Paanalahti
      • et al.
      Adverse events after manual therapy among patients seeking care for neck and/or back pain: a randomized controlled trial.
      ,
      • Haldeman S
      • Kohlbeck F
      • Mc Gregor M
      Risk factors and precipitating neck movements causing vertebrobasilar artery dissection after cervical trauma and spinal manipulation.
      ,
      • Haldeman S
      • Kohlbeck FJ
      • McGregor M.
      Stroke, cerebral artery dissection, and cervical spine manipulation therapy.
      For non-serious AEs, no other risk factors besides having experienced >60 days of neck pain in the preceding year (which may predict new or increased headache after spinal manipulation) are identified.
      • Rubinstein SM
      • Knol DL
      • Leboeuf-Yde C
      • van Tulder MW.
      Benign adverse events following chiropractic care for neck pain are associated with worse short-term outcomes but not worse outcomes at three months.
      So, in the absence of identified risk factors for non-serious AEs, how are we able to properly inform the patient about the risk? And, MT as applied in daily clinical practice rarely consists of HVT manipulation as a single modality, but rather tends to consist of a combination of interventions. With non-serious AE being common, it is hypothesized that AE events are not exclusively related to cervical manipulative thrust techniques.
      A treatment modality approach is needed in order to assess the occurrence of AE in MT interventions as applied in usual care manual therapy treatment and to accurately assess treatment modalities in relation to the occurrence of AE.
      The main aim of the current study is to gain insight in the occurrence of serious and non-serious AEs after MT treatment in patients with neck pain. Secondly, we want to explore the risk factors of non-serious AEs after spinal HVT manipulation and non-thrust mobilization in patients with neck pain.

      Methods and materials

      Design

      A prospective multicenter inception cohort study with a 12-month follow-up period in a Dutch manual therapy setting.

      Subjects

      Manual therapists

      All manual therapists participating in this study were licensed and registered. Participating manual therapists studied manual therapy for 3 or 4 years and hold a Master's degree in Manual therapy. Prior to the current study, all participants followed a 2-day course on the study protocol.

      Patients

      Patients entering the manual therapy practice with nonspecific neck pain, aged between 18 and 80 years and able to read and write Dutch were eligible for participation. Neck pain was defined as pain in the area between the occiput and the spine of scapulae.
      • Guzman J
      • Hurwitz EL
      • Carroll LJ
      • et al.
      A new conceptual model for neck pain: linking onset, course, and care: the Bone and Joint Decade 2000-2010 Task Force on neck pain and its associated disorders.
      Patients with known specific causes of neck pain (eg, known vascular or neurologic disorders, neoplasms, rheumatic conditions) were excluded. All patients received information on the study and signed an informed consent form. The study was approved by the medical ethics committee of Erasmus MC, Rotterdam, the Netherlands.

      Procedures

      Patient characteristics and clinical characteristics were registered by the MT on a website provided by the researchers. After registration, a unique code was assigned to each patient in order to combine treatment and AE data. Participating patients signed informed consent. Patients were screened for eligibility using the a priori defined criteria.

      Manual therapists

      Within a 3-month inclusion period, each manual therapist was asked to include 5 consecutive patients with non-specific neck pain. Manual therapists were instructed to provide usual care MT.
      The cervical spine was defined as the region between C0 and C6, whereas treatment on C7-T1 and treatment in the thoracic spine was excluded for analysis. Treatment modality, date, and number of sessions, as well as the process of clinical reasoning were registered by the therapists. Treatment modalities were a priori divided into 4 groups: HVT manipulation, non-thrust mobilization, a combination of HVT manipulation and non-thrust mobilization, and a group of other modalities. When a patient was manipulated within a treatment session, this session was defined as a manipulation (MP) session. When a patient received a mobilization technique, this session was defined as a mobilization (MOB) session (by the absence of manipulation). When the patient was treated with both manipulation and mobilization within 1 session, this was defined as MP+MOB session. Treatment sessions in which the patient received techniques other than manipulation or mobilization were defined as “other”. Besides these a priori categorized modalities, other modalities within the definition of “usual care” for patients with neck pain, like information, advice, or exercise can be provided.

      Patients

      Patients were asked to provide information on possible risk factors for AE. Reported risk factors for serious AE include age, biological sex, smoking, and presence of comorbidity (hypertension, [history of] heart failure, diabetes mellitus, hypercholesterolemia, atherosclerosis, and migraine).
      • Paanalahti
      • et al.
      Adverse events after manual therapy among patients seeking care for neck and/or back pain: a randomized controlled trial.
      ,
      • Haldeman S
      • Kohlbeck F
      • Mc Gregor M
      Risk factors and precipitating neck movements causing vertebrobasilar artery dissection after cervical trauma and spinal manipulation.
      ,
      • Haldeman S
      • Kohlbeck FJ
      • McGregor M.
      Stroke, cerebral artery dissection, and cervical spine manipulation therapy.
      At baseline, patients’ health status was surveyed using questionnaires on functioning, disability, and related factors, like psychological factors. Disability was measured using Neck Disability Index (NDI, range 0-50). Psychological factors included anxiety, depression, fear for (re)injury and activity related pain. Anxiety and depression were identified by the mean score of the anxiety and the depression questions of the Neck Bournemouth Questionnaire (NBQ, range 0-10). Avoidance of physical activities was measured with the the Fear Avoidance Beliefs Questionnaire- Physical Activity subscale (FABQ-PA, range 0-24).
      Additionally, based on the literature, a number of prognostic factors were measured: pain intensity (by means of a numeric rating scale for pain [NRS, range 1-10]), duration of neck pain, recurrence of neck pain, marital status, work status, sports engagement, and the presence of concomitant symptoms (eg, low back pain).
      The clinical reasoning and applied interventions were registered by the manual therapists. Patients reported on AE using the Adverse Events Questionnaire, which was filled out by the patient within 48 hours after each treatment. The Adverse Events Questionnaire asks if the patient experienced any unpleasant reaction after treatment, and requests to report the type of reaction, time of onset (range 1-4), duration (range 1-4), and intensity of symptoms (range 1-10). The questionnaire has been used before
      • Cagnie B
      • Vinck E
      • Beernaert A
      • Cambier D.
      How common are side effects of spinal manipulation and can these side effects be predicted?.
      ,
      • Rubinstein SM
      • Leboeuf-Yde C
      • Knol DL
      • De Koekkoek TE
      • Pfeifle CE
      • Van Tulder ME.
      The benefits outweigh the risks for patients undergoing chiropractic care for neck pain: a prospective, multicenter, cohort study.
      ,
      • Paanalahti
      • et al.
      Adverse events after manual therapy among patients seeking care for neck and/or back pain: a randomized controlled trial.
      and consists of several potential/possible post treatment reactions. Patients had the option to add reactions to the questionnaire that were not mentioned in the questionnaire.
      At the end of the treatment episode, patients filled out the NRS, NDI, NBQ, and FABQ again. Additionally, the Global Perceived Effect (GPE) questionnaire was filled out by the patients). The GPE is scored on a 7-point Likert scale ranging from “total recovery” to “worse than ever”. Test-retest reliability of GPE is excellent.
      • Kamper SJ
      • Ostelo RWJG
      • Knol DL
      • Maher CG
      • de Vet HCW
      • Hancock MJ.
      Global perceived effect scales provided reliable assessments of health transition in people with musculoskeletal disorders, but ratings are strongly influenced by current status.
      All questionnaires were returned to the researchers in a pre-stamped envelope. Design and timeline of the study are presented in figure 1.
      Fig 1
      Fig 1Timeline and design of the study. CR, clinical reasoning; TM, treatment modalities; T=0+n, end of treatment episode; T=0+12 months, follow-up.

      Data analysis

      Data were analyzed using the Statistical Package for the Social Sciences (SPSS) software (version 24.0). Four types of treatment modalities were defined (MP, MOB, MP+MOB, and other). Each type of treatment modality was analyzed separately for the risk of occurrence of AE. Additionally, patient characteristics were included in order to assess the influence of these patient characteristics on the relation between treatment modality and AE. Descriptive statistics (frequencies, mean, standard deviation, and range) were calculated for AE within each treatment group.

      Descriptives

      For all participating manual therapists, frequencies, mean, and standard deviation were calculated for age, years of work experience, weekly hours of work, and weekly number of patients with neck pain. Distribution of biological sex was registered.
      For all participating patients, frequencies for age, biological sex, duration of complaints, recurrence of complaints, direct access or referral by general practitioner or medical specialist, smoking, marital status, work status, sport engagement, and presence of concomitant symptoms were calculated.
      Within the treatment groups, frequencies of possible risk factors (age, biological sex, smoking, presence of comorbidity, pain intensity, disability, fear avoidance, and anxiety and depression) were calculated.

      Regression analysis

      Direct logistic regression analysis was performed to assess the effect of a number of factors on the likelihood that patients would experience an AE after manual therapy treatment. In order to grade AE, we decided that the occurrence of cramps, dizziness, blurred vision, nausea, tinnitus, vomiting, vertigo, and weakness of the limbs are undesirable and uncommon and the occurrence (intensity >1) can be defined as non-serious AE. For the commonly occurring headache, stiffness, aggravation of complaints, radiating pain, and fatigue, the intensity should be >5 and the duration longer than 24 h to be defined as a non-serious AE (figure 2). To prevent overfitting, the number of preselected predictors was chosen taking into account the 1:10 rule.
      • Steyerberg EW
      • Eijkemans MJ
      • Harrell Jr, FE
      • Habbema JD.
      Prognostic modelling with logistic regression analysis: a comparison of selection and estimation methods in small data sets.
      In order to check for multicollinearity, Pearson's correlation coefficients were calculated. In the case of significant correlation, regression analysis was performed with only 1 factor at a time. Factors were entered into logistic regression models for the different treatment groups (MP, MOB, MP+MOB, and “other modalities”). To describe possible risk factors of AE after manual therapy treatment, the following dichotomous independent variables were included into the model: biological sex, smoking, and presence of comorbidity. Continuous independent variables put into the model were age, pain intensity, disability, anxiety and depression, and fear avoidance. Occurrence of non-serious AE was included into the model as a dichotomous dependent variable.
      Fig 2
      Fig 2Timeline and design of the study. TIA, transient ischemic attack.

      Results

      Description of the study sample

      Manual therapists

      Of the 287 manual therapists eligible to participate, 263 joined the study. Reasons for non-participation (n=24) were pregnancy (n=2), working in a rehabilitation center (n=2), working in a foreign country (n=2), study delay/missed study deadline (n=16), and stopped studying (n=2). Most participating manual therapists were male, aged ≥40 years, with about 20 years of working experience. The average number of included patients per MT was 5. MT characteristics are presented in table 1.
      Table 1Manual therapist characteristics (n=263)
      Characteristic
      Age (y), mean ± SD42.2 (8.4)
      Male, n (%)207 (79%)
      Work experience (y), mean ± SD19.3 (7.1)
      Weekly h of work, mean ± SD24.6 (10.2)
      Weekly number of patients with neck pain, mean ± SD12.2 (8.0)

      Patients

      During the recruitment period, 3813 patients were eligible for participation; of these 1195 were included and 2618 were not. Main reasons for non-enrolment were that the MT already included the maximum of 5 consecutive patients, or patients were not willing to participate. 57.4% complete cases (consisting of both patient and manual therapist data) were received, resulting in 686 patients to be analyzed. Participants were predominantly female (69.4%) and 44.7% suffered from neck pain ≥12 weeks, mostly recurrent. Most patients had 1 or more concomitant complaints. Participants had an average age (SD) of 46.4 (13.6) years. Average pain score (SD) was 4.9 (2.1) on NRS (range 1-10). Mean disability measured on NDI (range 0-50) was 22.7 (6.4), and fear avoidance measured on FABQ (range 0-96) was 10. Anxiety and depression measured on NBQ (range 0-20) was 4.1 (2.3). Almost 1 in 4 patients were smokers (23.6%). In 11.4% of patients, comorbidity, in most cases hypertension (4.1%) and (history of) heart failure (2.3%), was present. Patient characteristics are shown in table 2.
      Table 2Patient characteristics
      VariablesStudy PopulationSampleNon-enrolled
      Patient characteristics (n= sample)n=1193n=686n=2618
      Biological sex (n=681), female (%)823 (69.4%)473 (69.4%)1636 (63.2%) (n=2587)
      Age, y (n=669), mean ± SD44.7 (13.7)46.4 (13.6)44.9 (16.6) (n=1856)
      Duration of NP, wk (n=613) (%)
       0-6420 (39.2%)259 (42.3%)
       6-12138 (12.9%)80 (13.1%)
       >12513 (47.9%)274 (44.7%)
      Recurrent neck pain (n=644), y (%)755 (66.9%)447 (69.3%)
      Marital status (n=659): Married (%)889 (77.2%)473 (77.1%)
      Work status (n=666); Employed (%)897 (77.1%)491 (73.7%)
      Smoking (n=682), y (%)300 (25.2%)161 (23.6%)
      Practicing sports (n=682), y (%)783 (65.9%)451 (66.1%)
      Concomitant symptoms (n=684)
      The total of this item is ≥100% because patients could indicate more than 1 area of concomitant symptoms.
      Total: 2190/ 1832
       Headache n (y, %)681 (57.0%)402 (58.8%)
       Low back pain n (y, %)448 (37.5%)313 (45.8%)
       Irradiating arm pain n (y, %)460 (38.5%)307 (44.9%)
       Disturbed sleep n (y, %)293 (24.5%)198 (28.9%)
       Concentration problems n (y, %)195 (16.3%)129 (18.9%)
       Memory problems n (y, %)113 (9.5%)81 (11.8%)
      NRS now (n=1183), mean (SD, range)4.8 (2.1, 1-10)4.9 (2.1, 1-10)
      NDI (n=1096) mean (SD, range)13.0 (6.5, 0-42)12.7 (6.4, 0-39)
      NBQ Anxiety and depression (n=1184) mean (SD, range)3.7 (2.6, 0-10)4.1 (2.3, 0-11)
      Note. Number in brackets in the 'variables' column are the number of people we have data available for in the sample.
      low asterisk The total of this item is ≥100% because patients could indicate more than 1 area of concomitant symptoms.

      Treatments

      In total, 3171 treatments were provided to the participating patients. Based on treatment modality, treatments were divided into 4 groups: MP, MOB, MP+MOB, and “other”. Every treatment session with its AE were analyzed separately. Patients’ characteristics corresponding with the information on treatment modalities and AE were taken into the analysis. Distribution of characteristics within the 4 treatment modality groups is shown in table 3.
      Table 3Characteristics of patients receiving different treatment modalities
      Manipulation (MP)Mobilization (MOB)MP+MOBNo MP nor MOB
      Age, y45.6 (13.2)47.9 (13.6)48.2 (13.4)46.1 (13.6)
      Biological sex, female69.7%68.3%64.6%70.6%
      Smoking, yes23.4%25.5%22.2%26.9%
      Comorbidity
      • Hypertension4.0%4.2%4.5%3.6%
      • (History of) heart failure2.3%2.8%2.8%2.3%
      • Hypercholesterolemia0.8%0.8%0.3%1.1%
      • Migraine1.7%1.1%1.4%1.8%
      • Diabetes1.0%1.1%0.7%1.4%
      • Multiple comorbidity1.5%2.2%2.4%2.1%
      • Absence of comorbidity88.7%87.7%87.8%87.7%
      NRS4.8 (2.1)5.0 (2.1)4.8 (2.1)4.9 (2.1)
      Range1-101-91-101-10
      NDI12.5 (6.2)13.1 (6.4)12.6 (6.2)13.0 (6.3)
      Range0-390-310-390-36
      FABQ10.3 (6.2)10.7 (6.4)10.6 (6.3)10.7 (6.1)
      Range0-280-240-240-24
      NBQ4.2 (2.3)4.2 (2.3)4.0 (2.3)4.2 (2.3)
      Range0-9.50-9.50-9.50-9.5

      Frequencies of non-serious adverse events

      No serious AEs were reported within our study.
      Within the MP group (in total 1014 sessions), stiffness was the most commonly reported AE (after 58.7% sessions), followed by headache (52.0%), radiation (50.3%), aggravation of complaints (48.2%), and fatigue (45.2%). Reports of vomiting (0.3%), disorientation/confusion (5.3%), and cramps (6.8%) were uncommon after manipulation.
      Within the MOB group (in total 829 sessions), stiffness was the most commonly reported AE (after 64.7% sessions), followed by radiation (64.0%), aggravation of complaints (57.0%), fatigue (55.6%), and headache (54.6%). Reports of vomiting (2.2%), disorientation/confusion (7.5%), and blurred vision (12.4%) were uncommon after mobilization.
      Within the MP+MOB group (in total 437 sessions), stiffness was the most commonly reported AE (after 62.0% sessions), followed by radiation (56.1%), headache (49.0%), aggravation of complaints (48.3%), and fatigue (44.6%). Reports of vomiting (0.7%), disorientation/confusion (6.4%), and blurred vision (8.9%) were uncommon after MP+MOB.
      Within the other modalities group (in total 891 sessions), stiffness was the most commonly reported AE (after 57.1% session), followed by radiation (55.8%), headache (54.0%), aggravation of complaints (50.4%), and fatigue (49.4%). Reports of vomiting (0.6%), disorientation/confusion (7.4%) and blurred vision (9.4%) were uncommon after treatment with other modalities.
      Frequencies of AE within the 4 treatment groups are presented in table 4.
      Table 4Descriptives of the occurrence of non-serious AE within treatment groups
      Manipulation (MP) (1014 sessions)Mobilization (MOB) (829 sessions)MP+MOB (437 sessions)“Other modalities” (891 sessions)
      % of treatments after which AE occursMean intensity (1-10) (SD) range reported% of treatments after which AE occursMean intensity (1-10) (SD) range reported% of treatments after which AE occursMean intensity (1-10) (SD) range reported% of treatments after which AE occursMean intensity (1-10) (SD) range reported
      Headache52.0%4.6 (2.0) 2-1054.6%4.6 (2.0) 2-1049.0%4.4 (1.9) 2-1054.0%4.6 (1.9) 2-9
      Radiation50.3%4.3 (1.6) 2-1064%4.8 (1.8) 2-956.1%4.6 (1.8) 2-1055.8%4.6 (1.9) 2-10
      Aggravation of complaints48.2%4.6 (1.9) 2-1057%4.9 (1.9) 2-948.3%4.8 (1.9) 2-1050.4%4.6 (1.8) 2-10
      Fatigue45.2%4.5 (1.8) 2-955.6%4.7 (2.0) 2-944.6%4.8 (1.9)49.4%4.8 (1.9) 2-9
      Stiffness58.7%4.3 (1.6) 2-1064.7%4.4 (1.8) 2-962.0%4.4 (1.8) 2-1057.1%4.3 (1.8) 2-10
      Cramps6.8%4.3 (2.0) 2-913.7%4.1 (1.7) 2-88.9%4.1 (1.4) 2-712.0%3.9 (1.7) 2-9
      Dizziness24.5%3.9 (1.8) 2-838.9%4.0 (1.8) 2-924.0%4.1 (1.9) 2-926.2%4.2 (1.8) 2-9
      Blurred vision8.7%3.7 (1.5) 2-812.4%3.8 (1.6) 2-78.9%3.6 (1.6) 2-89.4%4.3 (2.0) 2-8
      Nausea11.1%4.3 (1.7) 2-823.2%3.7 (1.5) 2-812.4%4.1 (1.5) 2-812.3%4.4 (1.7) 2-8
      Tinnitus7.3%1.2 (0.7) 1-619.9%3.6 (1.6) 2-711.0%3.4 (1.3) 2-610.6%3.5 (1.6) 2-7
      Vomiting0.3%5.0 (1.7) 4-72.2%3.6 (1.9) 2-70.7%5.0 (1.7) 3-60.6%4.0 (1.3) 3-6
      Limb weakness19.6%4.1 (1.7) 2-925.8%4.5 (2.1) 2-1022.7%4.1 (1.8) 2-821.4%4.2 (2.0) 2-10
      Disorientation/Confusion5.3%3.5 (1.4) 2-97.5%3.7 (1.9) 2-96.4%3.4 (1.7) 2-97.4%4.9 (2.2) 2-9
      NOTE. AE is defined as intensity >1. Dark gray indicating the highest prevalence rates, light gray indicating second highest prevalence rates.
      Differences in the occurrence of AE between groups were calculated for all treatments using chi square test. A statistical significant difference (χ2 (3, n=38,046)=46.7248, P<.001) in occurrence was found, detrimental to mobilizations (table 5).
      Table 5Chi square test between treatment groups
      MPMOBMP+MOBNo MP nor MOB ‘Other’Total
      X
      • Hurwitz EL.
      Epidemiology: spinal manipulation utilization.
      (3, n= 38,046)= 46.7248
      Present n (%)1162 (10%)1224 (12.3%)547 (10.4%)1217 (11.4%)4150
      Absent n (%)11,003 (90%)8723 (87.7%)4697 (89.6%)9473 (88.6%)33,896
      P<.001Total
      Total within all 13 AE.
      12,165 (100%)9947 (100%)5244 (100%)10,690 (100%)38,046
      NOTE. Differences (%) in the occurrence of AE between treatment groups for all treatments.
      low asterisk Total within all 13 AE.

      Risk factors of non-serious adverse events

      Binary logistic regression was performed to assess the effect of factors on the likelihood that patients would report an AE. Odds ratios (ORs) greater than 2.0 or smaller than 0.5 were considered to be clinically relevant. ORs are shown in table 6.
      Table 6Outcome of logistic regression analysis
      Factors Entering the ModelManipulation (MP) (1014 sessions)Mobilization (MOB) (829 sessions)MP+MOB (437 sessions)No MP nor MOB (891 sessions)
      CommonOR (95% CI)OR (95% CI)OR (95% CI)OR (95% CI)
      HeadacheAge, y
      Continuous variables.
      .97 (.95-.99).98 (.96-.99).95 (.91-.98).97 (.95-.99)
      Pain intensity
      Continuous variables.
      1.32 (1.09-1.61)1.31 (1.09-1.58)
      Smoking (yes)
      Dichotomous variables.
      3.33 (1.83-5.94)2.39 (1.35-4.24)
      Disability
      Continuous variables.
      1.09 (1.04-1.13)1.12 (1.06-1.19)1.07 (1.03-1.11)
      Anxiety + depression
      Continuous variables.
      1.24 (1.06-1.45)
      RadiationBiological sex
      Dichotomous variables.
      .65 (.42-.99)1.57 (1.03-2.38)
      Age, y
      Continuous variables.
      1.01 (1.0-1.03)
      Smoking (yes)
      Dichotomous variables.
      2.79 (1.91-4.08)1.57 (1.05-2.36)2.84 (1.58-5.12)2.86 (1.96-4.18)
      Pain intensity
      Continuous variables.
      1.26 (1.12-1.42)1.22 (1.02-1.45)1.16 (1.04-1.30)
      Presence of comorbidity
      Dichotomous variables.
      2.32 (1.22-4.44)
      Anxiety + depression
      Continuous variables.
      1.11 (1.003-1.23)
      Fear Avoidance
      Continuous variables.
      1.08 (1.03-1.34)
      Disability
      Continuous variables.
      .96 (.94-.98)
      Aggravation complaintsSmoking (yes)
      Dichotomous variables.
      1.7 (1.11-2.58)
      Disability
      Continuous variables.
      1.03 (1.01-1.05)1.02 (1.001-1.05)
      Presence of comorbidity
      Dichotomous variables.
      3.88 (1.62-9.26)
      Fear avoidance
      Continuous variables.
      .96 (.93-.99)
      Anxiety + depression
      Continuous variables.
      1.20 (1.08-1.34)1.16 (1.04-1.29)1.31 (1.18-1.46)
      StiffnessAge, y
      Continuous variables.
      .97 (.95-.98)
      Biological sex
      Dichotomous variables.
      .22 (.11-.46).28 (.095-.85).26 (.10-.66)
      Smoking (yes)
      Dichotomous variables.
      3.10 (1.83-5.27)2.19 (1.3-3.66)
      Pain intensity
      Continuous variables.
      1.26 (1.08-1.48)1.22 (1.05-1.43)1.18 (1.01-1.38)
      Presence of comorbidity
      Dichotomous variables.
      3.65 (1.26-10.58)
      FatigueBiological sex
      Dichotomous variables.
      .58 (.35-.95).52 (.31-.89)
      Age, y
      Continuous variables.
      .98 (.97-.99)
      Smoking (yes)
      Dichotomous variables.
      2.10 (1.37-3.22)2.70 (1.7-4.30)2.70 (1.7-4.3)
      Pain intensity
      Continuous variables.
      1.16 (1.04-1.30)
      Disability
      Continuous variables.
      .95 (.93-.97).97 (.95-.99).96 (.94-.98)
      Presence of comorbidity
      Dichotomous variables.
      .39 (.21-.73).40 (.21-.73)
      Anxiety +depression
      Continuous variables.
      1.16 (1.04-1.30)1.14 (1.01-1.30)1.28 (1.12-1.46)1.28 (1.12-1.46)
      Uncommon
      CrampsBiological sex
      Dichotomous variables.
      .58 (.35-.95)
      Age, y
      Continuous variables.
      1.03 (1.01-1.06)
      Smoking (yes)
      Dichotomous variables.
      2.10 (1.37-3.22)1.79 (1.12-2.87)
      Pain intensity
      Continuous variables.
      1.16 (1.04-1.30)1.25 (1.08-1.44)
      Disability.95 (.93-.97)
      Fear avoidance
      Continuous variables.
      .97 (.93-1.17)1.06 (1.02-1.1)
      Anxiety + depression
      Continuous variables.
      1.16 (1.04-1.30)
      DizzinessBiological sex
      Dichotomous variables.
      .57 (.40-.81)
      Age, y
      Continuous variables.
      .98 (.97-.99)
      Disability
      Continuous variables.
      1.04 (1.02-1.06)1.02 (1.001-1.04)
      Anxiety+ depression
      Continuous variables.
      1.16 (1.06-1.26)1.13 (1.04-1.24)1.16 (1.01-1.32)1.19 (1.09-1.3)
      Presence of comorbidity
      Dichotomous variables.
      .55 (.35-.88)
      Blurred visionPain intensity
      Continuous variables.
      1.31 (1.04-1.65)1.25 (1.08-1.44)
      Disability
      Continuous variables.
      1.04 (1.01-1.07)
      NauseaAge, y
      Continuous variables.
      .98 (.97-.998)
      Anxiety + depression
      Continuous variables.
      1.26 (1.12-1.43)
      Biological sex
      Dichotomous variables.
      .49 (.28-.86).59 (.36-.96).30 (.14-.67).49 (.27-.87)
      Pain intensity
      Continuous variables.
      1.28 (1.12-1.46)
      Disability
      Continuous variables.
      1.1 (1.02-1.08)1.04 (1.01-1.07)
      Fear Avoidance.95 (.91-.98).97 (.93-.99)
      Presence of comorbidity
      Dichotomous variables.
      3.33 (1.30-8.55)2.59 (1.07-6.29)2.99 (1.16-7.73)
      TinnitusDisability
      Continuous variables.
      1.06 (1.02-1.10)1.05 (1.02-1.09)
      Biological sex
      Dichotomous variables.
      1.80 (1.21-2.69)3.51 (1.83-6.71)
      Age, y
      Continuous variables.
      1.03 (1.001-1.05)1.03 (1.01-1.05)
      VomitingAge, y
      Continuous variables.
      .90 (.81-.96)
      Disability
      Continuous variables.
      1.23 (1.07-1.40)1.52 (1.05-2.21)1.56 (1.12-2.18)
      VertigoAge, y
      Continuous variables.
      .97 (.96-.99).98 (.96-.99)
      Smoking (yes)
      Dichotomous variables.
      .25 (.09-.71).38 (.15-.99).38 (.15-.99).37 (.18-.77)
      Pain intensity
      Continuous variables.
      .79 (.67-.93)
      Disability
      Continuous variables.
      .96 (.93-.99)1.04 (1.01-1.08)
      Presence of comorbidity
      Dichotomous variables.
      .45 (.22-.91)
      Anxiety + depression
      Continuous variables.
      1.25 (1.06-1.47)1.27 (1.06-1.51)1.27 (1.06-1.51)1.42 (1.21-1.67)
      Weakness of the limbsBiological sex
      Dichotomous variables.
      .57 (.33-.97)
      Presence of comorbidity
      Dichotomous variables.
      .47 (.29-.76)
      Dichotomous variables.
      Pain intensity
      Continuous variables.
      1.18 (1.06-1.32)1.26 (1.08-1.48)
      Anxiety + depression
      Continuous variables.
      1.18 (1.08-1.31)1.22 (1.11-1.34)
      NOTE. Empty cells did not significantly contribute to the model NB. Only variables which significantly contribute to the model are presented, italicized values are defined as clinically relevant (OR >2 or OR <0.5).
      low asterisk Continuous variables.
      Dichotomous variables.
      Analysis revealed that after MP treatment (1014 sessions in total) several common AEs were to be predicted: headache, stiffness, aggravation, radiation, and fatigue. Headache is predicted by age, pain intensity, smoking, and mean disability. Smoking showed to be the strongest predictor of headache after manipulation (OR 3.33 [95% CI 1.83-5.94]). Stiffness was predicted by smoking and pain intensity. Strongest predictor was smoking (OR 3.10 [95% CI 1.83-5.27]). Radiation was predicted by biological sex, smoking, and mean disability. Smoking also proved to be the strongest predictor of radiation after manipulation, recording an OR of 2.79 [95% CI 1.91-4.08]). The strongest predictor of fatigue after manipulation is again smoking (OR 2.10 [95% CI 1.37-3.22]).
      Analysis revealed that uncommon AEs after manual therapy treatment are to be predicted too. The strongest predictor of cramps after manipulation is smoking (OR 2.1 [95% CI 1.37-3.22]). With regard to nausea after manipulation, biological sex was associated, reporting an OR of 0.49 (95% CI 0.28-0.86) indicating that men had 51% lower odds of experiencing nausea after treatment with manipulation than women. The presence of comorbidity was the strongest predictor of nausea after manipulation, reporting an OR of 3.33 (95% CI 1.30-8.55).
      A total of 829 treatments consisted of MOB. Analysis showed that after MOB, stiffness, aggravation of complaints, radiation and nausea were to be predicted from the factors included in the analysis. Stiffness was predicted by pain intensity and presence of comorbidity, recording an OR of 3.65 (95% CI 1.26-10.58). Biological sex was associated, reporting an OR of 0.22 (95% CI 0.11-0.46), indicating that men had 78% lower odds of experiencing stiffness after MOB than woman. For aggravation of complaints, the presence of comorbidity is a strong predictor, recording an OR of 3.88 (95% CI 1.62-9.26). Radiation is also predicted by the presence of comorbidity (OR 2.32 [95% CI 1.22-4.44]) as well as nausea after MB treatment (OR 2.59 [95% CI 1.07-6.29]). Smoking is inversely associated with vertigo after mobilization, recording an OR of 0.38 (95% CI 0.15-0.99).
      After treatment with MP+MOB within 1 session, radiation was reported as an AE and analysis revealed that it can be predicted by smoking (OR 2.84 [95% CI 1.58-5.12]). Analysis revealed that the occurrence of fatigue is predicted by smoking as well (OR 2.70 [95% CI 1.70-4.30]). Men have lower odds of reporting fatigue and stiffness after MP+MOB, with ORs of 0.39 (95% CI 0.21-0.73) and 0.28 (95% CI 0.095-0.85). The occurrence of tinnitus as an AE after treatment was predicted by female sex (OR 3.51 [95% CI 1.83-6.71]). The OR >1 indicates that men are more likely to report tinnitus after MP+MOB. Smoking is inversely associated with vertigo after MP+MOB, recording an OR of 0.38 (95% CI 0.15-0.99). With regard to nausea after MP+MOB, biological sex was associated, reporting an OR of 0.30 (95% CI 0.14-0.87), which indicates that men have lower odds of experiencing nausea after treatment with MP+MOB than women.
      Although AEs are common after treatment with MP or MOB, they frequently occur after treatment with “other modalities” as well. The occurrence of headache, stiffness, fatigue, and nausea were predicted by factors included in the analysis. For headache after treatment with other modalities, smoking is the strongest risk factor (OR 2.39 [95% CI 1.35-4.24]). The same goes for stiffness after treatment with other modalities (OR 2.19 [95% CI 1.30-3.66]). Men have lower odds of reporting stiffness than women (OR 0.26 (95% CI 0.10-0.66). Radiation and fatigue are also predicted by smoking ([OR 2.86 (95% CI 1.96-4.1]) and (OR 2.70 [95% CI 1.7-4.3])). Presence of comorbidity predicts nausea after treatment with “other modalities” (OR 2.99 [95% CI 1.16-7.73]). Presence of comorbidity is inversely associated with vertigo and weakness of the limbs after “other modalities”, recording ORs of 0.45 (95% CI 0.22-0.91) and 0.47 (95% CI 0.29-0.76). Smoking is inversely associated with the effects of vertigo after “other modalities”, recording an OR of 0.37 (95% CI 0.18-0.77).

      Discussion

      The aim of the current study was to gain insight into the occurrence of serious and non-serious AE after spinal manipulation and mobilization in patients with neck pain. Secondly, we wanted to explore the risk factors of AE after spinal manipulation and mobilization in patients with neck pain. Fortunately, no serious AEs were reported. One remarkable finding was that there seems to be a statistical difference in the occurrence of non-serious AE after HVT manipulation and non-thrust mobilization, detrimental to mobilizations. When the occurrence of AE after manipulations is compared with the occurrence of AE after a treatment with “other modalities” (eg, information and advice, massage, transcutaneous electro neuro stimulation, coaching, questionnaires), no differences in occurrence appear to be present.
      Another remarkable finding in our study was that smoking (OR ranges from 2.10 [95% CI 1.37-3.22] to 3.33 [95% CI 1.83-5.93]) and the presence of comorbidity (OR ranges from 2.32 [95% CI 1.22-4.44] to 3.88 [95% CI 1.62-9.26]) had the strongest associations with reporting AEs after manual therapy treatment.
      Although cervical spine manipulation was considered primarily to be related to AE, there seems to be a significant difference in the occurrence of non-serious AEs detrimental to non-thrust mobilizations in our cohort.
      The cervical spine was defined as the region between C0 and C6, whereas treatment on C7-T1 and treatment in the thoracic spine were excluded from analysis. Although manipulation of the thoracic spine is a commonly chosen treatment technique in patients with neck pain,
      • Huisman PA
      • Speksnijder CM
      • de Wijer A.
      The effect of thoracic spine manipulation on pain and disability in patients with non-specific neck pain: a systematic review.
      we excluded it from the analysis, since we were interested in the occurrence of AE after manual therapy treatment in the cervical spine.
      The number of AEs reported in our study is higher than in other studies.
      • Cagnie B
      • Vinck E
      • Beernaert A
      • Cambier D.
      How common are side effects of spinal manipulation and can these side effects be predicted?.
      ,
      • Rubinstein SM
      • Knol DL
      • Leboeuf-Yde C
      • van Tulder MW.
      Benign adverse events following chiropractic care for neck pain are associated with worse short-term outcomes but not worse outcomes at three months.
      In other studies, when more than 1 AE was reported, only the most severe or the longest lasting AE is chosen to be included in the analysis. In our study, we analyzed all reported AEs, which may have led to a higher prevalence rate. Comparison with other studies is difficult, because of differences in reporting.
      According to Cagnie et al,
      • Cagnie B
      • Vinck E
      • Beernaert A
      • Cambier D.
      How common are side effects of spinal manipulation and can these side effects be predicted?.
      we have chosen to classify a treatment session as an MP session when spinal manipulation was used, regardless of the number of manipulations within the session. Cagnie et al
      • Cagnie B
      • Vinck E
      • Beernaert A
      • Cambier D.
      How common are side effects of spinal manipulation and can these side effects be predicted?.
      found that the number of performed manipulations could not be associated with the occurrence of AE. The direction and precise location in which the manipulation has been applied might be of interest. Further analysis should reveal whether 1 or more directions or locations of the manipulation could increase the chance of experiencing an AE after treatment.

      Occurrence of adverse events

      No serious AEs were reported in this cohort study. It is, however, possible that some non-serious AE have persisted or aggravated after 48 h. No data are available about the continuation of these events after 48 h.

      Manipulation

      When both groups were compared, the MP group and the other treatment groups had similar prevalence rates of AE.

      Mobilization

      Previous studies have suggested that when comparing spinal manipulation with spinal mobilization, the treatment effect seems to be the same, but the presence of AE appears to be more common among patients treated with manipulation.
      • Paanalahti
      • et al.
      Adverse events after manual therapy among patients seeking care for neck and/or back pain: a randomized controlled trial.
      ,
      • Leaver AM
      • Maher CG
      • Herbert RD
      • et al.
      A randomized controlled trial comparing manipulation with mobilization for recent onset neck pain.
      ,
      • Kranenburg HA
      • Schmitt MA
      • Puentedura EJ
      • et al.
      Adverse events associated with the use of cervical spine manipulation or mobilization and patient characteristics: a systematic review.
      Little is published in literature about the occurrence of AE after mobilization. In our study, we found AE to occur more frequent after mobilization than after manipulation treatment. Because little is known about the etiology of AE, it is hard to explain the differences in occurrence of AE between manipulation and mobilization. Manipulation techniques have been under debate for their possible serious AEs. These serious AEs are rare, whereas the non-serious AEs are far more frequently reported. Vautravers and Maigne
      • Vautravers P
      • Maigne JY.
      Cervical spine manipulation and the precautionary principle.
      claim that cervical spine manipulations should be contra-indicated for patients who experience dizziness, nausea, or headache persisting for more than 2 days. Ignoring these AEs increases the likelihood of harming the patient. Based on the outcome of our study, the question is raised whether this applies to mobilization too.

      Manipulation and mobilization

      It could be hypothesized that when a patient receives MP as well as MOB within 1 treatment session, reported AE rates should be higher than with a single modality. Our study shows that this is not what happens in clinical practice. Reported non-serious AEs are high, but not higher than in the MOB group (table 5). A possible explanation for this phenomenon is that when 2 treatment modalities are applied within 1 treatment session, the number of applied techniques decreases. On the other hand, the number of applied techniques seems of no consequence.
      • Cagnie B
      • Vinck E
      • Beernaert A
      • Cambier D.
      How common are side effects of spinal manipulation and can these side effects be predicted?.

      Other modalities

      We have chosen to create 1 group of treatment consisting of other modalities than MP or MOB. This group serves as a “control group” and is representative based on the fact that it contains patient characteristics that are similar to the other treatment groups. The fact that the prevalence rates of AE in this group is as high as in the MP, MOB, and MP+MOB group (table 5) suggests that AE are common, no matter what modality is applied.

      Limitations of the study

      The results of our study should be viewed within the limitations of the study. From the 1193 patients enrolled in the study, only 686 (57.5%) returned a complete set of data before deadline. We choose to ask the patients to return their questionnaires and report AE to the research center without the interference of the MT, in order to avoid socially desirable answers and responder bias. This, however, caused loss of data.
      Given that it is a longitudinal cohort study of the occurrence of AE within usual care manual therapy, patients are not randomized. Patients’ characteristics may a priori have contributed to the clinical decision of the manual therapist to apply manipulation or mobilization or other modalities.
      There are different ways to classify AE. In another study, a classification based on the occurrence of neurologic symptoms vs non-neurologic symptoms was used.
      • Hurwitz EL
      • Morgenstern H
      • Vassilaki M
      • et al.
      Frequency and clinical predictors of adverse reactions to chiropractic care in the UCLA neck pain study.
      Little is known about the etiology of AE
      • Haneline MT
      • Lewkovich GN.
      A narrative review of pathophysiological mechanisms associated with cervical artery dissection.
      and therefore we choose not to classify based on whatever seems to be a neurologic AE, but choose only to classify based on the intensity and duration of the patients’ reported AEs.
      • Weissman JS
      • Schneider EC
      • Weingart SN
      • et al.
      Comparing patient-reported hospital adverse events with medical record review: do patients know something that hospitals do not?.
      ,
      • Basch E
      • Jia X
      • Heller G
      • et al.
      Adverse symptom event reporting by patients vs clinicians: relationships with clinical outcomes.
      The differences in presenting the outcome highlight the complexity of accurately reporting AE.
      • Carlesso L
      • Rivett D.
      Manipulative practice in the cervical spine: a survey of IFOMPT member countries.
      We have chosen to analyze from the perspective of the treatment modality and to include patient characteristics and other factors as covariates into the regression analysis. Because every treatment episode has been split up into separate treatments, patient characteristics may/can be present in MP group, as well as in MOB group, MP+MOB group, and the “other treatments”-group, which might lead to over-representation. We suggest the recommendation that future research should include a propensity-score analysis, so that possible confounding based on the overlap of patient characteristics is compensated for.
      In our study, we asked patients to report AEs within 48 h after treatment, to avoid having too big a time interval which could affect the accuracy of reporting. We unfortunately have no information on how long the AE lasted after 48 h.
      We did not include oral contraceptives in our study. Oral contraceptives are described as a risk factor for AE after spinal manipulative treatment. In the study of Cagnie et al,
      • Cagnie B
      • Vinck E
      • Beernaert A
      • Cambier D.
      How common are side effects of spinal manipulation and can these side effects be predicted?.
      the use of oral contraceptives did not show any difference in the type of AE.
      It is difficult to determine whether the AE that arose were caused by the treatment provided. AE could be caused by other activities, or factors not covered in the questionnaire such as pre-existing problems, lifestyle, environmental effects, or concurrent treatment by other health care professionals.
      • Cagnie B
      • Vinck E
      • Beernaert A
      • Cambier D.
      How common are side effects of spinal manipulation and can these side effects be predicted?.
      ,
      • Kerry R
      • Taylor A.
      Cervical spine pre-treatment screening for arterial dysfunction: out with the old, in with the new.
      The focus of this study was to explore the role of patient characteristics in the occurrence of (N)SAE after manual therapy treatment. More research is needed to investigate the dose-response relation between the different treatment modalities and the occurrence of (N)SAE.

      Conclusion

      Our analysis showed a statistically significant difference in the occurrence of non-serious AEs between treatments consisting of manipulation, mobilization, a combination of both or even “other modalities” detrimental to mobilizations. In our study, no serious AEs were reported, which underlines the rare occurrence of these serious AEs. Non-serious AEs in manual therapy practice are common and can be predicted by smoking (OR ranges from 2.10 [95% CI 1.37-3.22] to 3.33 [95% CI 1.83-5.93]) and the presence of comorbidity (OR ranges from 2.32 [95% CI 1.22-4.44] to 3.88 [95% CI 1.62-9.26]).

      References

        • Magarey ME
        • Rebbeck T
        • Coughlan B
        • Grimmer K
        • Rivett DA
        • Refshauge K.
        Pre-manipulative testing of the cervical spine review, revision and new clinical guidelines.
        Man Ther. 2004; 9: 95-108
        • Hurwitz EL.
        Epidemiology: spinal manipulation utilization.
        J Electromyog Kines. 2012; 22: 648-654
        • Gross AR
        • Miller J
        • D'Sylva J
        • et al.
        Manipulation or mobilisation for neck pain: a Cochrane review systematic review.
        Man Ther. 2010; 15: 315-333
        • Bronfort G
        • Evans R
        • Anderson AV
        • Svendsen KH
        • Bracha Y
        • Grimm RH.
        Spinal manipulation, medication, or home exercise with advice for acute and subacute neck pain: randomized trial.
        Ann Intern Med. 2012; 156: 1-10
        • Chaibi A
        • Russell MB.
        Manual therapies for cervicogenic headache.
        J Headache Pain. 2012; 13: 351-359
        • Cagnie B
        • Vinck E
        • Beernaert A
        • Cambier D.
        How common are side effects of spinal manipulation and can these side effects be predicted?.
        Man Ther. 2004; 9: 151-156
        • Ernst E.
        Vascular accidents after neck manipulation: cause or coincidence?.
        Int J Clin Pract. 2010; 64: 673-677
        • Rubinstein SM
        • Leboeuf-Yde C
        • Knol DL
        • De Koekkoek TE
        • Pfeifle CE
        • Van Tulder ME.
        The benefits outweigh the risks for patients undergoing chiropractic care for neck pain: a prospective, multicenter, cohort study.
        J Manipul Physiol Ther. 2007; 30: 408-418
        • Malone DG
        • Baldwin NG
        • Tomecek FJ
        • et al.
        Complications of cervical spine manipulation therapy: 5-year retrospective study in a single-group practice.
        Neurosurgical Focus. 2002; 13: 1-8
        • Thiel HW
        • Bolton JE
        • Docherty S
        • et al.
        Safety of chiropractic manipulation of the cervical spine: a prospective national survey.
        Spine. 2007; 32: 2375-2378
        • Dvorak J
        • Orelli FV.
        How dangerous is manipulation of the cervical spine?.
        Man Med. 1985; 2: 1-4
        • Carnes D
        • Mullinger B
        • Underwood M.
        Defining adverse events in manual therapies: a modified Delphi consensus study.
        Man Ther. 2010; 15: 2-6
        • Carlesso LC
        • Gross AR
        • Santaguida PL
        • Burnie S
        • Voth S
        • Sadi J.
        Adverse events associated with the use of cervical manipulation and mobilization for the treatment of neck pain in adults: a systematic review.
        Man Ther. 2010; 15: 434-444
        • Senstad O
        • Leboeuf-Yde C
        • Borchgrevnink C.
        Predictors of side effects to spinal manipulative therapy.
        JMPT. 1996; 19: 441-445
        • Boyle E
        • Côté P
        • Grier AR
        • Cassidy JD.
        Examining vertebrobasilar artery stroke in two Canadian provinces.
        Spine. 2008; 33: S170-S175
        • Cassidy JD
        • Boyle E
        • Côté P
        • et al.
        Risk of vertebral stroke and chiropractic care: results of a population-based case control and case cross over study.
        Spine. 2008; 33: S170-S175
        • Ernst E
        • Canter PH.
        A systematic review of systematic reviews of spinal manipulation.
        J R Soc Med. 2006; 99: 192-196
        • Ernst E
        • Posadzki P.
        Reporting of adverse effects in randomised clinical trials of chiropractic manipulations: a systematic review.
        NZMJ. 2012; 125: 87-140
        • Wynd S
        • Estaway M
        • Vohra S
        • Kawchuk G.
        Correction: the quality of reports on cervical arterial dissection following cervical spinal manipulation.
        PLoS ONE. 2015; 10e0130221
        • Kerry R
        • Taylor A.
        Cervical spine pre-treatment screening for arterial dysfunction: out with the old, in with the new.
        InTouch. 2014; 147: 10-14
        • Paanalahti
        • et al.
        Adverse events after manual therapy among patients seeking care for neck and/or back pain: a randomized controlled trial.
        BMC Musculoskelet Disord. 2014; 15: 77
        • Haldeman S
        • Kohlbeck F
        • Mc Gregor M
        Risk factors and precipitating neck movements causing vertebrobasilar artery dissection after cervical trauma and spinal manipulation.
        Spine. 1999; 24: 785-794
        • Haldeman S
        • Kohlbeck FJ
        • McGregor M.
        Stroke, cerebral artery dissection, and cervical spine manipulation therapy.
        J Neurol. 2002; 249: 1098-1104
        • Rubinstein SM
        • Knol DL
        • Leboeuf-Yde C
        • van Tulder MW.
        Benign adverse events following chiropractic care for neck pain are associated with worse short-term outcomes but not worse outcomes at three months.
        Spine. 2008; 33: E950-E956
        • Guzman J
        • Hurwitz EL
        • Carroll LJ
        • et al.
        A new conceptual model for neck pain: linking onset, course, and care: the Bone and Joint Decade 2000-2010 Task Force on neck pain and its associated disorders.
        J Manipul Physiol Ther. 2009; 32: S17-S28
        • Kamper SJ
        • Ostelo RWJG
        • Knol DL
        • Maher CG
        • de Vet HCW
        • Hancock MJ.
        Global perceived effect scales provided reliable assessments of health transition in people with musculoskeletal disorders, but ratings are strongly influenced by current status.
        J Clin Epid. 2010; 63: 760-766
        • Steyerberg EW
        • Eijkemans MJ
        • Harrell Jr, FE
        • Habbema JD.
        Prognostic modelling with logistic regression analysis: a comparison of selection and estimation methods in small data sets.
        Stat Med. 2000; 19: 1059-1079
        • Huisman PA
        • Speksnijder CM
        • de Wijer A.
        The effect of thoracic spine manipulation on pain and disability in patients with non-specific neck pain: a systematic review.
        Disabil Rehabil. 2013; 35: 1677-1685
        • Leaver AM
        • Maher CG
        • Herbert RD
        • et al.
        A randomized controlled trial comparing manipulation with mobilization for recent onset neck pain.
        Arch Phys Med Rehabil. 2010; 91: 1313-1318
        • Kranenburg HA
        • Schmitt MA
        • Puentedura EJ
        • et al.
        Adverse events associated with the use of cervical spine manipulation or mobilization and patient characteristics: a systematic review.
        Musculoskel Sci Pract. 2017; 28: 32-38
        • Vautravers P
        • Maigne JY.
        Cervical spine manipulation and the precautionary principle.
        J Bone Spine. 2000; 67: 272-276
        • Hurwitz EL
        • Morgenstern H
        • Vassilaki M
        • et al.
        Frequency and clinical predictors of adverse reactions to chiropractic care in the UCLA neck pain study.
        Spine. 2005; 30: 1477-1484
        • Haneline MT
        • Lewkovich GN.
        A narrative review of pathophysiological mechanisms associated with cervical artery dissection.
        J Can Chiropr Assoc. 2007; 51: 146-157
        • Weissman JS
        • Schneider EC
        • Weingart SN
        • et al.
        Comparing patient-reported hospital adverse events with medical record review: do patients know something that hospitals do not?.
        Ann Intern Med. 2008; 149: 100-108
        • Basch E
        • Jia X
        • Heller G
        • et al.
        Adverse symptom event reporting by patients vs clinicians: relationships with clinical outcomes.
        J Natl Cancer Inst. 2009; 101: 1624-1632
        • Carlesso L
        • Rivett D.
        Manipulative practice in the cervical spine: a survey of IFOMPT member countries.
        J Man Manip Ther. 2011; 19: 66-70