From: Spine (Phila Pa 1976). 2010 Jul 12.
Neck pain is one of the most common musculoskeletal disorders, with an estimated 1-year prevalence of 31.4% to 35.6% of adults in the general population. The course of neck pain is characterized by exacerbations and remissions and only a small part of the patients experience complete resolution of their symptoms within 1 year.
A substantial proportion of the neck pain patients will thus develop chronic neck pain. The definition of chronicity differs between studies, in terms of either 3 or 6 months duration of complaints. Nevertheless, the estimates of their 1-year prevalence in the general population are similar: 8.7% to 17.8% for the 3-months definition, and 8% to 13.8% for the 6-months definition.
An important question is whether we can identify patients at risk of persistent complaints at the first consultation with the physician, based on their personal characteristics? This information could be helpful for a physician to gain insight into the prognosis of an individual patient with neck pain, and would aid in informing patients more accurately about their expected prognosis. Furthermore, this would aid researchers in selecting patients at high risk in studies on prevention of chronic neck pain.
Some studies have been conducted based on characteristics associated with persistence of neck complaints in the general population. Characteristics identified as a predictor in more than 1 study were age, duration of complaints, previous episode of neck pain, pain intensity, physical functioning, and accompanying low back pain. However, none of the studies constructed a prediction model that quantifies prognosis.
Therefore, the purpose of this study was to develop and externally validate a prediction rule that estimates the probability of complaints persisting for at least 6 months in patients consulting their physician for nonspecific neck pain.
The authors developed and externally validated a score chart to estimate the probability of persistent complaints at 6 months follow-up for patients with nonspecific neck pain in primary care. The chart has a moderate discriminative ability, an adequate calibration, a good fit, and a low explained variation in both the development and external population. The score chart predicts significantly better for every patient if he/she will have persistent complaints, than estimates for the overall population.
As mentioned in the introduction section, there have been studies evaluating predictors for persistence of nonspecific neck complaints in the general population. Eight of the 11 literature-based candidate predictors were also identified as a predictor in the analysis. The direction of the associations of these predictors in the authors model (i.e., worse or better prognosis) is consistent with those found in the previous studies. Gender, physical functioning, and duration of complaints were not identified as a predictor in this study, which could be caused by their borderline prognostic value: these variables showed a significant association with the outcome in only 10% to 45% of the analyses performed in the previous studies. This “borderline behavior” of variables seems to be a common phenomenon for predictors in neck pain: none of the predictors, identified in this study or previous studies, consistently has a (major) impact on prognosis. Of the 6 variables that were added to the literature-based variables, only “radiation of pain” was identified as a predictor in this study. Four of these 6 variables had a significant univariable association with the outcome, but (strong) correlation with other variables is probably the reason that they did not end up in the multivariable model.
The authors are the first to develop a multivariable model that quantifies persistence of nonspecific neck complaints. This hampers the comparison with other studies. The perfect model has a sensitivity and specificity of 1.0. Lower values for sensitivity and specificity result in overestimation or underestimation of the risk for persistent complaints. Our model is not perfect, but does make it possible to identify groups of patients with a significantly higher risk of persistent complaints. This facilitates future research on the prevention of chronicity in populations at high risk of developing persistent complaints. The possible influence of treatment on prognosis in the overall population is evaluated by the authors in another study.
The clinical value of this study is that it provides a tool for physicians to estimate the risk of persistent complaints in patients with nonspecific neck pain in primary care, based on easily obtainable data at baseline. The discriminative ability is moderate, which is reflected by the small percentage of patients in the 2 extreme risk categories. However, the calibration and fit of the model are good, and the score chart shows significantly better for every patient if he/she will have persistent complaints, than estimates for the overall population. This makes it informative for patients and physicians. For researchers, identification of high risk groups facilitates research on effectiveness of interventions to prevent persisting complaints.
The authors developed and externally validated a score chart to estimate the probability of persistent complaints at 6 months follow-up for patients with nonspecific neck pain in primary care. The chart has a moderate discriminative ability, an adequate calibration, a good fit, and a low explained variation in both the development and external population. The score chart predicts significantly better for every patient if he/she will develop persistent complaints, than estimates for the overall population. This makes the score chart informative for patients and physicians and facilitates future research in patients with high risk of developing chronic neck pain.
Risk factors for chronic neck complaints include age, pain intensity, previous neck complaints, radiation of pain, accompanying low back pain, accompanying headache, employment status, health status, and cause of complaints.
We have a neck pain prognosis form developed with the parameters set forth in this article.
