Analysis of the validity of the cervical flexion rotation test in diagnosing cervicogenic headache | M A N U E L L T E R A P I شات الطائف - شات

Analysis of the validity of the cervical flexion rotation test in diagnosing cervicogenic headache

Arkivert i ukategorisert 0 kommentarer
Print Friendly

This post was eventually triggered by a youtube video where Doctor of Physical Therapy, Francois Prizinski, presents the cervical flexion rotation test as a “very powerful test” to diagnose cervicogenic headaches. This interpretation is based on a QUADAS score hitting 12 out of 14 possible points. That is truly, as Dr. Prizinski also accentuates, “extremely good”. He also introduces the test referring to sensitivity, specificity, positive likelihood ratios and other cool statistic terms giving the impression that the cervical flexion rotation test should be imperative in the screening of cervicogenic headache.

YouTube Preview Image

Honestly, my post is not really about the cervical flexion rotation test and I don’t really even have a strong opinion whether or not the test is a “must-do” in screening cervicogenic headaches.

However, I do have issues about how to really evaluate and present a diagnostic test based on evidence based quality assessment, such as ie. the QUADAS.

My issues about this is partly founded on the first essay I wrote as part of my MSc-degree where I evaluated the main studies regarding the cervical flexion rotation test as a screening tool for cervicogenic headaches. I learned three things from this. Firstly,  it obviously expanded my knowledge base on headaches in general. Secondly, it took me deep down into how to really evaluate statistical and methodological aspects of studies representing diagnostic tests. And it also gave me an understanding of how difficult it is to put a diagnostic label to a condition in the neuromusculoskeletal system.

I would say in general, that there seems to be a lot of focus on teaching this and that test to unveil this and that condition and students and practitioners will many times be quite pedantic about its truthfulness. I don’t really blame those who have arrived to their conclusions pointing to “best available evidence”, but sometimes I do believe that many have too much unsubstantiated trust in research.

The rest of this post is therefore a full copy of my essay about the cervical flexion rotation test and hopefully it will exemplify and highlight some important aspects I think one should consider when evaluating the basis of which biases and factors to look out for when any test represents a sky-high sensitivity and specificity. It does not cover all aspects of this process, but at least it was a pass…

1.    Introduction

The notion of the neck being a source of head pain has been subject for substantial controversy and scientific research for the last century (1). Despite increasing support of the nosography and pathogenesis of cervicogenic headache (2,3), there is still not a consensus on the question of whether or not the neck can constitute a key role in the initiation of a headache (1,4).

There was however, a turning point in 1983 when Sjaastad and his co-workers introduced the term “cervicogenic” headache (CeH), and the first proposal on a set of diagnostic criteria (5). The following awareness of CeH as a possible distinct entity consequently prompted The International Association for the Study of Pain (IASP) to accept CeH as a specific form of headache in 1994 (6). However, the IASP diagnostic criteria differ from those of the Cervicogenic Headache International Study Group (CHISG) and also from the most recent definition presented by the International Headache Society (IHS). The discrepancy of the diagnostic criteria can be attributed to the significant overlap of symptoms, which are commonly associated with other types of headaches such as migraine without aura and tension-type headache (7-10).

On the basis of subjective characteristics, CeH is in principle a predominantly unilateral headache that does not change side. The evidence of symptoms and signs of neck involvement are essential. Whilst the CHISG in their diagnostic criteria include reduced range of motion (ROM) of the neck and precipitation of head pain by neck movements or awkward head positioning, IHS points out that these clinical features are not unique to CeH (7). However, the IHS accentuates that identification of such clinical signs must demonstrate reliability and validity (8).

Considering impairments of cervical function is reported as a common association with CeH, the physical examination of the cervical spine has become an important part of the evaluation of the headache patient (10-15). As physiotherapists commonly encounter patients with varying headaches, it is crucial for the therapist to be able to distinguish between the headaches, as treatment is likely to be unsuccessful if misdiagnosed (11,12). Considering that an incorrect headache diagnosis may take place in 50% of the cases (16), there is the potential for many headache patients to receive inappropriate treatment, potentially prolonging and compromising the recovery the patient (11,12).

Recently there has been focus on the potential of the cervical flexion rotation test (FRT) to detect dysfunction of the C1-2 joint, and it has been proposed as an accurate and reliable tool in the assessment of CeH (17-19). For the physiotherapist who questions the validity of differential diagnosis of headaches, the availability of a reliable and valid diagnostic tool would be useful. Hence this essay aims to critically evaluate the studies by Hall et al (2004, 2008) and Ogince et al (2008) in an attempt to clarify the role of the FRT in diagnosis of CeH in terms of its validity.

2.    The rationale of the cervical flexion rotation test

The most accepted neurophysiological mechanism for CeH is the neural convergence of the upper three cervical spinal nerves and the spinal trigeminal nucleus (20,21). Any pathology of the structures innervated by the C1 to C3 spinal nerves can be potential sources for CeH, but the type and definite localisation of pathology varies considerably among individuals (3).

Preliminary work has shown that the C1-2 joint accounts for about 50% of the total cervical rotation (22,23), and that this amount of rotation is not affected by age or gender (25,25). As CeH is thought to have a primary involvement of the upper cervical segments in accordance with the convergence theory related to the trigeminocervical nucleus (2), it gives meaning to examine for dysfunction of the C1-2 joint as a possible cause of CeH. There is also research to indicate that manual examination by means of PPIVMs and PAIVMs can detect symptomatic cervical joint dysfunctions in the CeH subject (26-29), however Hall et al (2004, 2008) and Ogince et al (2008) remark that these tests involve a high degree of skill from the examiner and therefore proposes the cervical FRT as an easily applied tool to evaluate upper cervical joint impairment in CeH.

The intended purpose of applying cervical rotation in a position of full neck flexion as described by Dvorak et al (1984) is when pre-tensioning the structures above and below the C1-2 joint with full cervical flexion, movement from other segmental levels will be restrained and isolate the rotational movement to the atlanto-axial joint (30).

The physical performance of the test is well described in the papers by Hall et al (2004, 2008) and Ogince et al (2008), but it is worth noting that the FRT is performed in a supine position in these studies, in contrast to other authors that have positioned the subjects in sitting as originally described by Dr. Dvorak (15,28,30,31). However, the intention of the test is the same between the authors.

3.    Description of studies

Whilst there exists some evidence on the reliability of the FRT in evaluation of asymptomatic subjects (21,24,25), there is a scarcity of evidence on the FRT as a valid diagnostic tool in identifying CeH subjects. Table 1 shows the studies that will be evaluated with focus on validity of the FRT within this context.

Table 1. Synopsis of included studies.


Hall T and Robinson K (2004)

Objective: Investigate active cervical ROM and the FRT to compare ROM in CeH subjects with asymptomatic controls. Sample group: 28 subjects with CeH (mean age 43.3±11.5 years; 20 females) and 28 asymptomatic subjects (mean age 43±13 years; 20 females). Method: Two examiners assessed active cervical ROM and the FRT in both groups and noted ROM using the Cervical Range of Motion Device. Prior to assessment, one therapist determined the C1-2 as the dominant symptomatic cervical motion segment by means of PPIVMs and PAIVMs as the reference standard.

Ogince M, et al (2007)

Objective: Investigate sensitivity and specificity of the FRT in diagnosing C1-2 related CeH. Sample group: 23 subjects with CeH (mean age 46 years; 20 females), 23 asymptomatic controls (mean age 40 years; 15 females) and 12 subjects with migraine with aura (mean age 37 years; 3 females). Subjects ranged in age from 18 to 66 years. Method: Two examiners assessed FRT in all three groups and noted ROM using the Cervical Range of Motion Device. Identification of C1-2 as symptomatic segment was determined by PPIVMs and PAIVMs as the reference standard.

Hall T, Robinson K, et al (2008)

Objective: Investigate reliability and validity of the FRT in diagnosing C1-2 related CeH. Sample group and method: In study 1, two experienced examiners evaluated the FRT in 20 subjects with CeH (mean age 37±13 years; 14 females) with C1-2 as primary dysfunctional level, 10 asymptomatic controls (mean age38±13 years; 8 females) and 10 subjects with CeH (mean age38±13 years; 6 females), but without C1-2 as the primary dysfunctional level. In study 2, two inexperienced and 1 experienced examiner evaluated the FRT in 12 subjects with CeH (mean age 29±4 years; 12 females) and 12 asymptomatic controls (mean age 30±7 years; 7 females).

3.1.         Study design

The optimal structure of a study is one that is designed to minimize the possibility for bias (32). Each of the studies of Hall et al (2004, 2008) and Ogince et al (2008) has case control designs, a design that is suitable for exploratory studies as they can offer large amount of information over a short period of time from relatively few subjects (33). The opportunity of retrospective analysis of a potential relationship between C1-2 dysfunction and CeH can be used to offer evidence to support the hypothesis and justify subsequent clinical trials. Unfortunately, they also have several deceptive weaknesses, generally making them a low level source of evidence (34).

The most obvious disadvantage is their vulnerability to spectrum bias, as subjects sought out for the study sample may not be representative for a prospective patient population as they are included based on specific characteristics (35). A case control designed study faces a two-ended challenge early in the study, namely the identification of “cases” and the selection of “controls” where both ends are susceptible to biases that could potentially affect the accuracy and utility of the diagnostic test (33).

In both studies of Hall et al (2004, 2008), as well as in the study of Ogince et al (2008), the examiners were blinded to the subjects’ group allocation and from the findings of the other involved examiners, thus mitigating the risk of review bias (34).

3.2.         Study population

An inappropriate spectrum composition of subjects included to a study does not reflect the environment in which the test is applied clinically where patients represent with symptoms commonly confused with the condition of interest (36). Furthermore, the diagnostic accuracy of a test can be artificially overestimated if the diagnostic test under investigation is compared between a group of subjects identified to have the disease and a separate group of healthy control subjects (37).

Although Hall et al (2004) examined a CeH group against an equal number of asymptomatic subjects it does not automatically influence the internal validity of the study, but will confine its clinical relevance (38). It must however, be noted that out of the 53 subjects interviewed for this study, 28 were found suitable for inclusion as having CeH. In contrast, Ogince et al identified 46 out of 325 headache subjects as suffering from CeH. As they employed same inclusion and exclusion criteria, this clear discrepancy in prevalence between the groups suggest a confounding variable. Whilst the latter study identified CeH subjects through telephone interviews after recruitment from medical specialists and by advertisements, Hall et al (2004) offers no information on how the subjects were interviewed other than the subjects were recruited from private physical therapist clinics and advertisements in local newspaper. Varying skill on behalf of the interviewer(s) or lack of standardisation of the interview can provide some explanation to this as subtle changes in nonverbal signals, tone of voice or non-uniform wording of questions can significantly influence the response from the subjects and subsequently who enters the study (33).

Hall et al (2008) paralleled C1-2 related CeH subjects with an equal group of subjects with CeH that had a dysfunctional segment other than C1-2, and a third group of asymptomatic subjects. In this study, the subjects were recruited as a sample of convenience from medical and physiotherapy clinics and physiotherapy course participants. Although this method has its advantages in terms of time, cost and logistics, it has some potential for selection bias as it relies on a subjective decision from the investigator(s) on whom to participate or not (39). There is no clarification in the paper of how the subjects were recruited or who recruited them, nor is any justification for this method offered. The validated use of convenient sampling is generally when the research extends over longer durations to investigate seasonal changes or temporal variations important to the research question (33), but it is generally considered the method of lowest credibility (39).

When including only subjects with classical and severe symptoms of CeH, the authors engender spectrum bias, making the findings from the studies inapplicable to those with less severe symptoms (34). In the study of Ogince et al (2008), the control subjects suffering from migraine with aura were excluded if they experienced any neck pain, discomfort or stiffness. The contribution of these exclusion criteria remains unclear, but further research including both control groups with migraine with and without aura may contribute to external validity of the results from these studies.

In all three studies CeH subjects prior to examination filled out a headache questionnaire, which has shown to be reliable (40). This provided the authors with a demographic profile of headache severity based on duration, intensity and frequency of attacks. The information was used to expose a possible relationship between headache severity and potential restriction of the FRT. There is however no further sub-group analysis of the spectrum composition in terms of possible influencing factors such as pathology, clinical features or co-morbidity. Caution should therefore be exerted when interpreting these results, as there is research to suggest enough manifestation of physical differences between the traumatic and non-traumatic onset of CeH to classify CeH patients into two separate categories (4,41).

4.    Validity of the reference standard

Central to the authority of investigating any diagnostic test is the reference standard, which should have demonstrated validity to justify its use as a benchmark for comparison (32). To verify “true” dysfunction of the atlanto-axial joint, all three studies utilised PAIVMs as described by Maitland (42). Whilst the authors argue that these techniques have shown capacity in detecting symptomatic cervical joint dysfunction in CeH subjects in several studies (26-29), there are other scientific circles that refute the diagnostic validity of the same techniques (43-45). One possible reason for this disagreement is the current lack of a valid reference standard to observe a bona fide involvement of the facet joint. The only feature that is consistent within the diagnostic criteria as described by IASP, CHISG and IHS, is the demand for abolition of headache as a result from a positive anaesthetic blockade effect to a cervical structure or its nerve supply (7-9). The consensus for this criterion is odd in light of the fact that anaesthetic blockade injections to facet joints have failed to prove a consistent response, as well as lack of a reference standard comparison in the studies (45,46). Further research may help develop a reliable and valid reference standard to assess PAIVMs as diagnostic tools of spinal segmental dysfunctions in the future.

For many conditions, there are unfortunately no valid reference standard diagnostic test (47). One important principle for a credible reference standard is that there should preferably be no mechanistic relationship to the diagnostic test (34). This makes the reference standard as used in these studies vulnerable to incorporation bias (32). This bias makes the result of the FRT better than what it really is, because patients that might represent with slightly less severe cases were never included to the studies. The results are also accentuated by an element of verification bias, as the reference standard was not applied to any of the non-CeH subjects (33). Since negative findings with application of the FRT were never verified with the reference standard, it is not known if there occurred any true false-negatives in the control group.

5.    The scoring criteria

A positive test was based on an eyeballed estimation of more than 10° restriction of anticipated range of rotation, and was measured using a Cervical Range of Motion Device previously shown to be reliable (49). Whilst Hall et al (2004, 2008) also defined a positive test on pain provocation and resistance to movement during the FRT, whichever came first, Ogince et al (2008) deliberately excluded pain provocation during their procedure for reasons unknown.

In addition neither of the authors clarify if there is any significance on finding pain but not firm end feel or limitation of movement. Nor is it discussed if pain should be related to the neck or to the head or similar to familiar headache symptom. These aspects seems important as the CHISG diagnostic criteria stress precipitation of the familiar head pain occurring during neck movement or by external pressure over the cervical or occipital region on the symptomatic side as one of the sole criteria. Restriction of neck ROM is not (7). Hall et al (2004) found that headache index severity (40) was significantly correlated to degree of restriction of rotation, but Ogince et al found no such relationship. Hence the interpretation of the possible combinations of these parameters remains unclear.

Using the receiver operating characteristic (ROC) curve they calculated that less than or equal to 32° or 33° would be the cut-off value for a positive test or not. The clinical value of this is limited as CeH subjects are likely to present with continuity from mild to severe cases, and cannot easily be dichotomised as having CeH or not (50).

6.    Data analysis

The sensitivity and specificity of 91% and 90% (18), and 90% and 88% (19) respectively, are not easily extrapolated to clinical practice because they require a reverse analytical way of normal clinical reasoning from the therapist in weighing their relative values (51). Furthermore, they are likely to be inflated because of the highly defined C1-2 related CeH cases as a consequence of the studies’ selection bias (52).

Hall et al (2008) and Ogince et al (2008) reported diagnostic accuracies of 89% (18) and 91% (19), based on calculations from ROC curves. However, the pragmatic value of these findings are severely limited as ROC curves only demonstrate the mathematical relationship between sensitivity and specificity (53), which were likely to be overestimated, thus affecting the calculated diagnostic accuracy likewise.

The predictive values indicate the probability that subjects with a positive or negative test are indeed correctly diagnosed and are very useful for the clinician (54), but can be equally fallacious as the high prevalence of CeH in the sample group would likely result in corresponding high positive predictive values (PPV). Whilst Ogince et al (2008) reports an average PPV of 85.7%, the prevalence of C1-2 related CeH in the sample group gives a pre-test probability of 39.6% (19). Considering epidemiological studies of chronic and frequent headache sufferers have found a prevalence of CeH ranging from 15% to 18% (55-57), bearing in mind that the incidence of C1-2 related CeH within this group was not taken into account, the predictive values recorded by Ogince et al (2008) must be interpreted with caution.

Likelihood ratios, on the other hand, are not affected by the prevalence of disease (36). They are however, affected by the quality of the study that determined their value (54) and are therefore only valid in the same clinical situation as they were calculated in.

7.    Conclusion

Although there has been presented some evidence on the ability of the FRT to identify C1-2 dysfunction in CeH, the clinical value of these findings suffers from limitations in the general methodological quality of the studies by Hall et al (2004, 2008) and Ogince et al (2008).

First of al, even if the FRT appears to hold criterion-related validity, it is still not clear whether the FRT truly measures movement of the C1-2 motion segment, and even so, to which extent. The results do anyhow justify further research such as monitoring movement of the upper cervical joints by means of radiographic measurement while applying the FRT to elucidate this further. Secondly, the study design allowed only the CeH subjects to undergo the reference test. All of those in the migraine and asymptomatic control group who were reported with negative findings with the FRT could have been found to be positive on the reference test, thus opening the study for verification bias increasing the sensitivity of the test. Thirdly, the authors leaves no comments on the possibility for indeterminate test results, hence how to interpret a potential finding where FRT is painful or not in either side of the neck, only in the head or both or reproduces familiar pain and when limited with less than 10° is unknown.

It is important for the physiotherapist to acknowledge that even if these findings appear hold some internal criterion-related validity, the CeH subjects were highly defined with high pre-test probabilities because of the many levels of spectrum and selection bias, hence not representative for the normal clinical population. One could argue on the other hand, to determine if a diagnostic tool is able to distinguish disease from non-disease, the initial studies need to have such highly defined cases as in these studies. Caution must anyway be made when interpreting the FRT in the clinical setting as it has limited external validity. Further research is required to define the exact role and relationship between the FRT, C1-2 dysfunction and CeH.

8.    Self reflection

It is far beyond any doubt that the making of this essay has been an enormous and challenging task, both academically and personally for me as an international newcomer to the masters program. The assiduous undertaking of microscopic analysis and critical evaluation of our profession’s ability to diagnose the cervicogenic headache patient has definitively been a strenuous journey, and it rapidly made me appreciate just some of the immense amount of work and dedication it must lie within the scientists and authors who constantly feed new aspects of knowledge into our profession of neuromusculoskeletal practice.

Whilst writing this essay, I have attempted to investigate the research methods behind the high sensitivity, specificity and diagnostic accuracy reported by the authors that studied the validity of the FRT. To discover the multitude of controversy behind the theories of CeH, and then analyse the far-reaching consequences of the variety in clinical diagnostic criteria, made this investigation into a complex exercise. Orientating myself within the aspects of causes and effects of the many potential biases and levels of methodology within the case-controlled study design was at times frustrating, but also rewarded me with some awareness of the testimonial behind the making of a diagnostic test.

I was previously unaware of all the critical factors that needs to be taken into consideration when interpreting the value of a diagnostic test. This journey has made me reflect on not only how to apply the findings of the FRT to the headache patient, but also in respect to the other diagnostic tests I have commonly applied in my clinical practice. It has certainly been valuable in ways more than I expected when I first started on this essay.

In respect to my understanding of the FRT after this review, it has revealed certain levels of bias error that now makes it more challenging to incorporate the test the way I used to, as I now realise that I have attached more unsubstantiated credibility to the FRT than I should. This new knowledge allows me to contribute more to discussions regarding the diagnostic factors that need to be considered in encountering the CeH patient and issues related to validity of other diagnostic tests used in evaluation of the neuromusculoskeletal system.


  1. Antonaci F, et al. Concepts leading to the definition of the term cervicogenic headache: a historical overview. J Headache Pain 2005;6:462-466.
  2. Bogduk N. Cervicogenic Headache: Anatomic basis and Pathophysiological Mechanisms. Current Pain and Headache Reports 2001;5:382-386.
  3. Antonaci F, Bono G, Chimento P. Diagnosing cervicogenic headache. Journal of Headache and Pain 2006;7:145-148.
  4. da Silva H, Bordini C. Cervicogenic headache. Current Pain and Headache Reports 2006;10:306-311.
  5. Sjaastad O, Saunte C, Hovdal H, et al. Cervicogenic headache: an hypothesis. Cephalalgia 1983;3:249-256.
  6. Merskey H, Bogduk N (eds). Classification of chronic pain: description of chronic pain syndromes and definition of pain terms. IASP Press, Seattle, WA 1994.
  7. Sjaastad O, Fredriksen TA, Pfaffenrath V. Cervicogenic headache: diagnostic criteria. The Cervicogenic Headache International Study Group. Headache 1998;38:442-445.
  8. International Headache Society. The international classification of headache disorders 2nd edition. Cephalalgia 2004;24(Suppl. 1):9-160.
  9. van Suijlekom H, de Vet H, van den Berg S, Weber W. Interobserver reliability in physical examination of the cervical spine in patients with headache. Headache 2000;40:581-586.
  10. Biondi D. Cervicogenic Headache: A Review of Diagnostics and Treatment Strategies. The Journal of the American Osteopathic Association 2005;105(4 suppl.)S16-S22.
  11. Jull G. Management of cervical headache. Manual Therapy 1997;2(4):182-190.
  12. Hall T, Briffa K, Hopper D. Clinical evaluation of Cervicogenic Headache: A Clinical Perspective. The Journal of Manual & Manipulative Therapy 2008;16(2):73-80.
  13. Watson D, Trott P. Cervical headache: an investigation of natural head posture and upper cervical flexor muscle performance. Cephalalgia 1993;13:272-284.
  14. Stolk-Hornsveld F, et al. Impaired mobility of cervical spine as a tool in diagnosis of cervicogenic headache. Cephalalgia 1999:19:436-440.
  15. Zito G, Jull G, Story I. Clinical tests of musculoskeletal dysfunction in the diagnosis of cervicogenic headache. Manual therapy 2006;11:118-129.
  16. Pfaffenrath V, Kaube H. Diagnostics of cervicogenic headache. Functional Neurology 1990;5:159-164.
  17. Hall T, Robinson K. The flexion-rotation test and active cervical mobility – a comparative measurement study in cervicogenic headache. Manual Therapy 2004;9:197-202.
  18. Ogince M, Hall T, Robinson K, Blackmore A.M. The diagnostic validity of the cervical flexion-rotation test in C1/2 related cervicogenic headache. Manual Therapy 2007;12:256-262.
  19. Hall T, Robinson K, et al. Intertester reliability and diagnostic validity of the cervical flexion-rotation test. Journal of Manipulative and Physiological Therapeutics. 2008 May;31(4):293-300.
  20. Ashkenazi A, Silberstein S. Headache management for the pain specialist. Regional Anasthesia and Pain Medicine 2004;29(5):462-475.
  21. Amiri M, Jull G, Bullock-Saxton J. Measuring range of active cervical rotation in a position of full head flexion using the 3D Fastrak measurement systsem: an intra-tester reliability study. Manual Therapy 2003;8(3):176-179.
  22. Panjabi M, et al. Three-dimensional movements of the upper cervical spine. Spine 1988;13:726-731.
  23. Hiroshi I, et al. Three-dimensional movement analysis of the upper cervical spine during axial rotation. Spine 1993;18:2388-2392.
  24. Dvorak J, Antinnes J, Panjabi M, Loustalot D, Bonomo. Age and gender related normal motion of the cervical spine. Spine 1992;17(10):S393-S398.
  25. Smith K, Hall T, Robinson K. The influence of age, gender, lifestyle factors and sub-clinical neck pain on the cervical flexion-rotation test and cervical range of motion. Manual Therapy 2008;13:552-559.
  26. Treleaven J, Jull G, Atkinson L. Cervical musculoskeletal dysfunction in post-concussional headache. Cephalalgia 1994;14:273-279.
  27. Hanten W, Olson S, Ludwig G. Reliability of manual mobility testing of the upper cervical spine in subjects with cervicogenic headache. The Journal of Manual and Manipulative Therapy 2002;10(2):76-82.
  28. Jaeger B. Are cervicogenic headaches due to myofascial pain and cervical spine dysfunction? Cephalalgia 1989;9:157-164.
  29. Schoensee H, et al. The effect of mobilisation on cervical headaches. Journal of Orthopaedic and Sports Physical Therapy 1995;21:181-196.
  30. Dvorak J, Dvorak V, Schneider W. Manual Medicine, 1st ed. Berlin, Heidelberg: Springer;1984(Chapter 2):70-79.
  31. Amiri M, et al. Cervical musculoskeletal impairments in frequent intermittent headache. Part 2: Subjects with concurrent headache types. Cephalalgia 2007;27:891-898.
  32. Fritz M, Wainner R. Examining diagnostic tests: An evidence-based perspective. Physical Therapy 2001;81(9):1546-1564.
  33. Hulley S, et al. Designing clinical research 3rd ed. Philadelphia: Lippincott Williams & Williams; 2007.
  34. Mayer D. Essential evidence-based medicine. Cambridge: Cambridge University Press; 2004.
  35. Reid M, Lachs M, Feinstein A. Use of methodological standards in diagnostic test research: getting better but still not good. Journal of American Medicine Association 1995;274:645-651.
  36. Jeaschke R, Guyatt G, Sackett DL. Users’ guides to the medical literature, III: how to use an article about a diagnostic test, A: Are the results of the study valid? Journal of American Medicine Association 1994;271:389-391.
  37. van der Schouw Y, Verbeek A, Ruijs S. Guidelines for the assessment of new diagnostic tests. Investigative Radiology 1995;30:334-340.
  38. Kelly S et al. The identification of bias in studies of the diagnostic performance of imaging modalities. The British Journal of Radiology 1997;70:1028-1035.
  39. Patton, M. Qualitative evaluation and research methods. SAGE Publications; Newbury Park London New Delhi 1990.
  40. Niere K, Robinson P. Determination of manipulative physiotherapy treatment outcome in headache patients. Manual Therapy 1997;2:199-205.
  41. Dumas J, et al. Physical impairments in cervicogenic headache: traumatic vs. nontraumatic onset. Cephalalgia 2001;21:884-893.
  42. Maitland G, Hengeveld L, Banks K, English K. Maitland’s vertebral manipulation 6th ed. Oxford: Butterworth-Heinemann; 2001.
  43. Troyanovich S, Harrison D. Motion palpation: it’s time to accept the evidence. Journal of Manipulative & Physiological Therapeutics 1998;21(8):568-571.
  44. King W, et al. The validity of manual examination in assessing patients with neck pain. The Spine Journal 2007;7:22-26.
  45. Bogduk N, McGuirk B. Cervicogenic headache: diagnosis (chapter 2). Management of acute and chronic neck pain. Elsevier 2006.
  46. Nordin M et al. Assessment of neck pain and its associated disorders. Results of the Bone and Joint Decade 2000-2010 Task Force on Neck Pain and its associated disorders. Journal of Manipulative and Physiological Therapeutics 2009;32(Suppl. 2):S117-S140.
  47. Carragee E, Haldeman S, Hurwitz E. The pyrite standard: the Midas touch in the diagnosis of axial pain syndromes. The Spine Journal 2007;7:27-31.
  48. Greenhalgh T. How to read a paper: papers that report diagnostic or screening tests. BMJ 1997;315:540-543.
  49. Rheault W, Albright B, et al. Intertester reliability of the cervical range of motion device. Journal of Orthopaedic and Sports Physical Therapy 1992;15(3):147-150.
  50. Grimes D, Schulz K. Uses and abuses of screening tests. The Lancet 2002;359:881-884.
  51. Gallagher E. The problem with sensitivity and specificity. Annals of Emergency Medicine 2003;42:298-303.
  52. Lijmer J, Mol B, Heisterkamp S, et al. Empirical evidence of design-related bias in studies of diagnostic tests. Journal of American Medicine Association 1999;282:1061-1066.
  53. Cook N. Use and misuse of the Reciever Operating Characteristic Curve in Risk Prediction. Circulation 2007;115:928-935.
  54. Altman D, Bland M. Diagnostic tests 2: Predictive values. British Medical Journal 1994;309(6947):102 (9 July).
  55. Nilsson N. The prevalence of cervicogenic headache in a random population sample of 20-59-year olds. Spine 1995;20:1884-1888.
  56. Stovner L, Hagen K, Jensen R, et al. The global burden of headahce: A documentation of headache prevalence and disability worldwide. Cephalalgia 2007;27:192-210.
  57. Sjaastad O, Bakketeig L. Prevalence of cervicogenic headache: Vågå study of headache epidemiology. Acta Neurologica Scandinavia 2008;117:173-180.
Postet av Sigurd   @   22 February 2012 0 kommentarer
Tags : , , , , ,

0 kommentarer

Ingen kommentarer enda. Bli den første til å legge igjen en kommentar!
Legg igjen en kommentar

Forrige post
Neste post
EXULT designed by ZENVERSE  |  In conjunction with Reseller Hosting from the #1 Web Hosting Provider - HostNexus.