The aim of this study is to compare Dutch usual care musculoskeletal therapy in patients with non-specific neck pain with recommendations from international clinical practice guidelines. Physical therapy is diverse, as it may consist of exercise, massage, advice, and other modalities. Physical therapists with post graduate qualifications in manual therapy (MT) may additionally apply spinal thrust manipulation or non-thrust mobilization techniques to treat neck pain. It is important that, in the absence of a Dutch clinical guideline for the treatment of patients with neck pain, musculoskeletal therapists use the available recommendations from international clinical practice guidelines when treating patients with neck pain. One updated clinical practice guideline was identified (Blanpied, 2017), a report from the Task Force on Neck Pain (Guzman et al., 2008) and the IFOMPT International Framework for Examination of the Cervical Region for potential of Cervical Arterial Dysfunction prior to Orthopaedic Manual Therapy Intervention (Rushton et al., 2014). At baseline 1193 patients are included and data with regard to pain, disability, fear avoidance, expectations and applied treatment modalities are gathered. Outcome is measured using the Global Perceived Effect questionnaire. Results show that patients with acute neck pain are treated significantly more often with manipulation compared to patients with sub-acute or chronic neck pain (p < .000) and younger patients are treated with manipulation more often than older patients (p < .000). In the presence of comorbidity, the preference of spinal manipulation seems to diminish, in favour of mobilization and exercise. Almost every patient receives multimodal therapy (94.3%) and spinal manipulation and mobilization are rarely used as a stand-alone treatment (4.5% and 0.8%). Dutch musculoskeletal therapists choose treatment strategies that correspond with recommendations from international guidelines.
Dutch MTs follow recommendations from international guidelines.
Acute NP is more often treated with manipulation than chronic NP (p < .000).
Younger patients receive manipulations more often than older patients (p < .000).
High velocity-thrust manipulation is applied frequently (33.8%).
Patients’ with NP are commonly treated with multimodal therapy (94.3%).
Neck pain is a common condition with peak prevalence rates up to 6.5% in Western Europe, high costs due to disability, and high recurrence rates ( ). Physical therapy is a common treatment option for patients with acute, sub-acute, and chronic non-specific neck pain. Physical therapy is diverse, as it may consist of exercise, massage, advice, and other modalities. Physical therapists with post graduate qualifications in manual therapy (MT) may additionally apply spinal thrust manipulation or non-thrust mobilization techniques to treat neck pain.
Evidence-based clinical guidelines are systematically developed statements that aim to enhance the quality of treatments and to provide healthcare professionals with evidence-based recommendations to assist clinical decision-making and reduce variability in clinical practice and benefits to patients ( ; ).
With regard to the treatment of patients with nonspecific neck pain, one international clinical practice guideline is available ( ). This guideline was updated in 2017, by Blanpied and colleagues. The guideline recommends clinicians to consider the use of cervical spinal thrust manipulation and mobilization procedures to reduce neck pain. Based on strong evidence, the guideline states that combining cervical spinal manipulation and mobilization with exercise is more effective for reducing neck pain, headache, and disability than spinal manipulation and mobilization alone ( ). Besides this guideline, other studies underline the added value of multimodal (e.g. spinal manipulation and exercise) treatment of patients with neck pain ( ; ; ). For instance, for non-specific neck pain without trauma, recommendations from this report are that spinal manipulation, mobilization, supervised exercises, MT (existing of spinal manipulation, mobilization, massage) plus exercises are likely helpful and worth considering ( ).
The duration of neck pain may play an important role in the choice of treatment. For instance, psychosocial factors are an important prognostic indicator of prolonged disability since they contribute to the transition of an acute condition to a chronic, disabling condition. When psychosocial factors are identified and patients suffer from chronic neck pain, the guideline advises the approach to be modified to emphasize active rehabilitation and less ‘hands-on’ therapy ( ; ).
Although evidence supports the effectiveness of spinal manipulative therapy, its use remains a matter of debate because of the adverse events potentially associated with spinal manipulative therapy. As a result, a consensus document from the International Federation of Orthopaedic Manipulative Physical Therapists (IFOMPT) was published ( ). MT interventions for the cervical spine addressed in this framework include spinal manipulation, mobilization, and exercise. The aim of the framework is to provide guidance for the assessment of the cervical region for potential cervical artery dysfunction (CAD) prior to MT intervention. Arterial dissection and other vascular presentations (in their early presentation) may mimic musculoskeletal dysfunction ( ) but are fairly recognisable if the appropriate questions are asked during patient history taking. Besides proper history taking, the application of MT should be based on a comprehensive assessment of the patient’s neuro-musculoskeletal system and of the patient’s functional abilities. An important underlying principle of the framework is that the physical therapist cannot rely on the results of one single test to draw conclusions. A combination of sources of information (history taking, risk factor analysis, pre-manipulative testing) should be used to distinguish those conditions that are indications or contra-indications to MT and/or demand special precautions, as well as those where anatomical anomalies or pathological processes limit or direct the use of MT ( ; , ).
The framework considers the following risk factors to be associated with an increased risk of CAD: past history of cervical spine trauma, history of migraine-type headache, hypertension, hypercholesterolemia, cardiac disease, vascular disease, previous cerebrovascular accident or transient ischaemic attack, diabetes mellitus, blood clotting disorders, anticoagulant therapy, long-term use of steroids, history of smoking, recent infection, immediately post-partum, recent head or neck trauma and the absence of a plausible mechanical explanation for the patient’s symptoms ( ). Evidence shows that general cardiovascular risk factors, with the exception of migraine, do not appear to be important risk factors for CAD, but appear to be risk factors for other vascular pathology such as atherosclerosis ( ; ; ). Risk factors for atherosclerosis are often present in older people and thrombotic stroke is typically a disease of the elderly ( ; ). Therefore, age is an important factor in the consideration of applying spinal manipulation because atherosclerotic related factors may be present.
In The Netherlands, physical therapy including spinal manipulation and mobilization is a common treatment option for patients with neck pain, but an evidence-based clinical guideline with regard to the treatment of nonspecific neck pain is lacking. More importantly, it is unknown what comprises physical therapy treatment for patients with neck pain in The Netherlands in the absence of a guideline. Therefore, the aim of the current study is to describe usual care physical therapy in primary care in patients with non-specific neck pain and to compare this with international clinical guidelines and recommendations. Secondly, we aimed to evaluate whether treatment modalities vary in subgroups based on age, duration of complaints, and the presence of risk factors.
A prospective multi-center cohort study with 12-months follow-up was conducted in a primary care physical therapy setting in the Netherlands. The Medical Ethical Committee of the Erasmus MC Rotterdam approved this study. A complete description of the design is published elsewhere.
All participants in this study were physical therapists with postgraduate qualifications in manual therapy (MTs) and registered with the Royal Dutch Society for Physical Therapy (KNGF). They all worked in a primary or secondary healthcare setting. MTs completed data collections as a part of the requirement of their Masters’ research course. The MTs were asked to provide usual care MT. Prior to the study, all participants followed a two-day course which explained the registration method of the study. To prevent from bias, the comparison with international guidelines was unmentioned and the stated aim of the study was to gain insight into the ‘black box’ of usual care MT. Consequently, no effort was made to standardize the number or the content of the treatment sessions. Within a three-month inclusion period, the participating MTs included five consecutive patients with non-specific neck pain.
Patients with non-specific neck pain ( ; ) were eligible. Inclusion criteria were non-specific neck pain with or without other complaints (e.g. headache, radiating symptoms) and age between 18 and 80 years. Patients were excluded if there were signs of major pathology, if patients were not able to read and write Dutch language, and if they suffered from specific neck pain due to severe trauma, systemic disorders, or generalized neurologic syndromes. All patients signed an informed consent prior to enrolment in the study.
Self-administered questionnaires were used to collect data from all included patients. During the first consultation, the patient signed an informed consent form and filled out baseline questionnaires. To be able to merge baseline and follow-up data and to ensure anonymity, MTs provided all questionnaires with a unique code. During the treatment episode, the MTs described treatment dates, their process of clinical reasoning, and the modalities applied in a free text box for each treatment. Finally, the reason to stop the therapy was noted. All questionnaires of patients and MTs were returned to the research center using pre-stamped envelopes. A research assistant called the MTs with missing questionnaires within 3 weeks after the turn in deadline. Fig. 1 presents the study design.
The baseline questionnaires included patients’ characteristics (age, gender) and neck pain characteristics (recurrence, localisation, duration of complaints, risk factors, comorbidity). Validated questionnaires were filled out: Numeric Rating Scale (NRS) for pain, functioning by means of the Neck Disability Index (NDI) and the Neck Bournemouth Questionnaire (NBQ), and the Fear Avoidance Beliefs Questionnaire (FABQ). At baseline, patients’ expectations regarding recovery were assessed. At the end of the treatment episode, patients’ recovery was assessed using the Global Perceived Effect (GPE) scale.
Descriptive statistics (SPSS 24.0) were used to present baseline characteristics of patients and data on treatments applied by the MTs. All authors attended to a consensus meeting where it was determined that recommendations to be followed were: whether multimodal therapy is the preferred treatment, whether patients with chronic pain are treated with a hands-off approach and whether in the presence of risk of adverse events (factors to be measured: age, hypertension, hypercholesterolemia, a history of heart failure and migraine). Chi-square tests were performed to assess for difference in the applied treatment modalities between subgroups based on age, duration, and the presence of risk factors. To prevent from bias, no participating therapists were in attendance.
Duration of complaints was categorized into: acute neck pain (0–6 weeks), sub-acute neck pain (7–12 weeks), and chronic neck pain (>12 weeks). Age was dichotomized into younger (<45 year) and older (≥45 year) based on the occurrence of CAD ( ).
Patients’ reported risk factors as described in the IFOMPT screening document included in our study were hypertension, hypercholesterolemia, a history of heart failure and migraine ( ; ; ; ). Patients with diabetes mellitus, long term steroid use, anti-coagulant therapy, recent infection, immediately post-partum, or multi-morbidity were assigned to the multiple comorbidity group.
Based upon treatment categories in the literature, treatment modalities were categorized into spinal manipulation, mobilization, exercise, information and advice, and ‘other’ modalities applied to the cervical spine.
Of the 287 MTs eligible to participate, 263 joined the study and enrolled patients during the recruitment period. Reasons for dropout of MT’s were: pregnancy (n = 2), working in a rehabilitation center (n = 2), working in a foreign country (n = 2), delay in Master’s program/missed deadlines (n = 16), stopped Master’s program (n = 2). The majority (n = 207, 79%) of the MTs were male. Mean age of participants was 42.2 (8.4) years and mean working experience was 19.3 (7.1) years ( Table 1 ).
|Manual Therapists’ characteristics|
|Age (years), mean (SD)||42.2 (8.4)|
|Male, n (%)||207 (79%)|
|Work experience (years), mean (SD)||19.3 (7.1)|
|Weekly hours of work, mean (SD)||24.6 (10.2)|
|Weekly number of neck pain patients, mean (SD)||12.2 (8.0)|
The MTs recruited 1311 patients with non-specific neck pain, of which 1193 generated baseline data. During the recruitment period, all patients with neck pain who were eligible but refused to participate (for whatever reason) or were supernumerary because the MT already included 5 patients, were registered and information on age and gender was collected. This information was used to check the representativeness of the study group. Student T-test showed no differences regarding age and gender between enrolled and non-enrolled patients with neck pain. Participants were predominantly female. Two-third of patients had recurrent neck pain and the majority of patients had one or more concomitant complaints. Almost half of the population at baseline experienced neck pain for more than 12 weeks. With regard to concomitant complaints, 2190 complaints were reported by 1096 patients. Main reported concomitant symptoms were headache (681 patients, 62.1%), low back pain (448 patients, 40.1%), and radiating arm pain (460 patients, 41.9%) ( Table 2 ). Unfortunately, not every MT was able to include 5 patients. Main reasons given were: unpredicted absence of the MT during the inclusion period (e.g. illness), not enough eligible patients and postal failure. A number of questionnaires were returned to the education center without the required (anonymized) patient details, which should have been filled out by the MT in order to merge data.
|Study population||Non- participants|
|Patients’ characteristics||Baseline (n = 1193)||Baseline (n = 2618)|
|Age (n = 1170), mean (sd)||44.7 (13.7)||44.9 (16.6) (n = 2587)|
|Gender (n = 1186), female (%)||823 (69.4%)||1636 (63.2%) (n = 1856)|
|Duration (n = 1072) (%)|
|Acute 0–6 wks||420 (39.2%)|
|Sub-acute 6–12 wks||138 (12.9%)|
|Chronic >12 wks||513 (47.9%)|
|Concomitant symptoms: n = 1096|
|Total reported: 2190**|
|Headache yes (%)||681 (62.1%)|
|Low back pain yes (%)||448 (40.1%)|
|Radiating pain yes (%)||460 (41.9%)|
|Disturbed sleep yes (%)||293 (26.7%)|
|Concentration problems yes (%)||195 (17.8%)|
|Memory loss yes (%)||113 (10.3%)|
|Recurrent neck pain (n = 1129) yes (%)||755 (66.9%)|
|Marital status (n = 1152) yes (%)||889 (77.2%)|
|Work status (n = 1164) yes (%)||897 (77.1%)|
|Smoking (n = 1190) yes (%)||300 (25.2%)|
|Medication use (n = 1189) yes (%)||560 (47.1%)|
|Sports (n = 1189) yes (%)||783 (65.9%)|
|Previous experience manual therapy (n = 1169) yes (%)||407 (31.0%)|
|NRS now (n = 1183), mean (sd, range)||4.8 (2.1; 1–10)|
|NRS min (n = 1183) mean (sd, range)||3.7 (2.0; 1–10)|
|NRS max (n = 1183) mean (sd, range)||6.6 (2.2; 1–10)|
|NDI (n = 1096) mean (sd, range)||13.0 (6.5, 0–42)|
|NBQ (n = 1171) mean (sd, range)||28.3 (12.9; 0–68)|
|FABQ (n = 1053) mean (sd, range)||26.6 (16.6; 0–85)|
|FABQ-W (n = 1103) mean (sd, range)||13.4 (12.2; 0–60)|
|FABQ-PA (n = 1129) mean (sd, range)||13.2 (7.3; 0–30)|
|PEL Expected recovery due to:|