Exploring Patient Satisfaction: A Secondary Analysis
of a Randomized Clinical Trial of Spinal Manipulation,
Home Exercise, and Medication for Acute and Subacute Neck Pain

This section is compiled by Frank M. Painter, D.C.
Send all comments or additions to:

FROM:   J Manipulative Physiol Ther 2014 (Sep 5);   37 (8):   593–601 ~ FULL TEXT

Brent D. Leininger, DC, Roni Evans, DC, PhD, Gert Bronfort, DC, PhD

Research Fellow,
Integrative Health & Wellbeing Research Program,
Center for Spirituality & Healing,
University of Minnesota,
Minneapolis, MN.

OBJECTIVE:   The purpose of this study was to assess satisfaction with specific aspects of care for acute neck pain and explore the relationship between satisfaction with care, neck pain, and global satisfaction.

METHODS:   This study was a secondary analysis of patient satisfaction from a randomized trial of spinal manipulation therapy (SMT) delivered by doctors of chiropractic, home exercise and advice (HEA) delivered by exercise therapists, and medication (MED) prescribed by a medical doctors for acute/subacute neck pain. Differences in satisfaction with specific aspects of care were analyzed using a linear mixed model. The relationship between specific aspects of care and (1) change in neck pain (primary outcome of the randomized trial) and (2) global satisfaction were assessed using Pearson's correlation and multiple linear regression.

RESULTS:   Individuals receiving SMT or HEA were more satisfied with the information and general care received than MED group participants. Spinal manipulation therapy and HEA groups reported similar satisfaction with information provided during treatment; however, the SMT group was more satisfied with general care. Satisfaction with general care (r = –0.75 to –0.77; R2 = 0.55–0.56) had a stronger relationship with global satisfaction compared with satisfaction with information provided (r = –0.65 to 0.67; R2 = 0.39–0.46). The relationship between satisfaction with care and neck pain was weak (r = 0.17–0.38; R2 = 0.08–0.21).

CONCLUSIONS:   Individuals with acute/subacute neck pain were more satisfied with specific aspects of care received during spinal manipulation therapy or home exercise interventions compared to receiving medication. The relationship between neck pain and satisfaction with care was weak.

Key Indexing Terms:   Neck Pain, Patient Satisfaction, Musculoskeletal Manipulations, Exercise Therapy, Pharmaceutical Preparations, Clinical Trial, Chiropractic

From the FULL TEXT Article:


Neck pain is one of the most commonly reported health complaints in primary care settings. [1, 2] As concern for costs and side effects related to treating spinal pain conditions continues to grow, the search for effective, patient-centered treatments has become paramount. Patient satisfaction has become a widely advocated means for measuring patients' preferences and views related to treatment quality in clinical practice. [3] Furthermore, it is recommended as a core outcome domain for chronic pain clinical trials by the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials group. [4]

A large percentage of health care visits are made to medical doctors, doctors of chiropractic, and physical therapists, who use a range of interventions to manage neck complaints. [5] Although commonly used for the management of acute or subacute neck pain, systematic reviews have found only limited to low-quality evidence for spinal manipulation, exercise, and medications. [6–8] Recently, in one of the first large randomized trials investigating spinal manipulation therapy (SMT) for acute and subacute neck pain, our group found that patients receiving SMT experienced significantly greater reductions in pain than those receiving medication in the short and long terms. [9] No significant group differences were found between SMT and home exercise for most outcomes, including pain. An exception was global satisfaction, with the SMT group significantly more satisfied compared with the home exercise and medication groups, and the home exercise group more satisfied than those who received medication. The satisfaction-related findings, while secondary, are potentially important, especially given the lack of research that is currently available examining patient satisfaction in the existing acute neck pain literature. [6] Furthermore, it is not known if there were specific aspects of care that informed patients' global satisfaction and if these differed by treatment. Such insights may provide important information that can affect the implementation of research findings into clinical practice and the design of future patient-centered research.

Although satisfaction outcomes are growing in popularity, recent studies and commentaries have questioned the interpretation of patient satisfaction and its utility in both health care and clinical research settings. [10, 11] Fenton et al [10] examined the relationship between patient satisfaction, health care expenditures, and health using the Medical Expenditures Panel Survey. Surprisingly, they found that increased satisfaction is associated with higher medical expenditures and mortality. Similar but less extreme findings are emerging in the spinal pain literature, which find improved patient satisfaction with increased diagnostic tests and treatment, regardless of clinical outcomes. [12, 13]

In light of the emerging questions about utility of satisfaction as an outcome measure and findings from our recent study, [9] we sought to further explore the patient satisfaction domain. The purpose of this article was to assess

(1) treatment group differences in satisfaction with specific aspects of care in acute neck pain patients receiving SMT, home exercise, and medication as measured by a multidimensional satisfaction questionnaire and

(2) the relationship between specific aspects of satisfaction with care and both change in neck pain (primary outcome measure in parent randomized clinical trial) and global satisfaction (secondary outcome).


Although patient satisfaction is a commonly promoted outcome measure, [4, 18] its utility and interpretation has become somewhat controversial. [10, 11] The results of these secondary analyses provide a better understanding of the issues related to satisfaction with 3 common treatment approaches for acute neck pain, providing additional context for interpreting the primary study's findings and gleaning a better understanding of how patients experienced the interventions. [9]

The primary results of our previously reported randomized clinical trial found that SMT patients experienced significantly greater pain reduction than MED patients; SMT patients also reported greater global satisfaction than HEA, and both groups were more globally satisfied than the MED group. [9] Through secondary analyses exploring specific aspects of satisfaction, we confirmed a consistent advantage for SMT and HEA over MED in terms of satisfaction in subdomains related to general care (which included provider concern, quality of treatment recommendations, and overall care) and information provided (including cause, prognosis, activities to hasten recovery, and prevention). Patients receiving SMT were also more satisfied with general care than the HEA group. The secondary analyses also revealed the HEA group to be as satisfied with information received as the SMT group. The HEA group was most satisfied with specific information related to activities to hasten recovery and prevent future neck pain. These findings, although not entirely surprising given the information-rich nature of the HEA intervention, are noteworthy. Systematic reviews and recent qualitative work have found that patients with spinal pain appreciate information regarding the cause of their symptoms and ways to manage their condition; furthermore, they are frustrated by the lack of such information in clinical encounters. [19, 20] The results of our secondary analyses suggest that HEA as an intervention might offer an advantage in meeting these types of informational needs for acute neck pain patients and should be considered more frequently in clinical practice and future research.

We also found acute neck pain patients' global satisfaction to be largely explained by general care– and information-related factors (eg, 60% of the variance) but not entirely. Although the findings suggest that these factors are important, they also suggest that patients consider additional satisfaction-related factors not included in the multidimensional satisfaction instrument used in this study. Indeed, other studies have found that “process related factors” (eg, how care is delivered) including treatment format and the nature of the patient-provider interaction as well as outcomes are important domains considered by patients when assessing their satisfaction. [21–25] Future qualitative and quantitative studies exploring the full range of satisfaction-related factors that inform acute neck pain patients' preferences and perspectives, including their relationship to other important outcomes such as health care utilization and costs, are needed.

Somewhat surprising is our finding that change in pain is poorly explained by satisfaction with general care and information in this acute neck pain population (21% of variance). This finding confirms the observations of others that satisfaction with treatment should not be directly equated with effectiveness (as defined by impact on pain, disability, and other important outcomes). [10–12] Rather, satisfaction with care is perhaps better viewed as the patients' reflection on the treatment experience, illustrating a range of perspectives regarding what they actually receive, how it is delivered, and how they experienced it. [21]

This study also highlights general issues regarding how to best measure patient satisfaction. One commonly used approach (used in the primary study from which this work is derived [9]) is to use a global scale of satisfaction. While appealing in its simplicity, the global nature of such scales can mask individual satisfaction-related factors such as the ones identified in this secondary analysis. Furthermore, it is possible for patients to be satisfied with some aspects of care and dissatisfied with others, which will not be clearly identified with a global measure. Incomplete understanding of what aspects of care patients find satisfying and dissatisfying can play an important role in patients' ability and willingness to engage in specific treatments. Consequently, multidimensional evaluations of satisfaction should be considered for researchers and clinicians desiring to fully understand the patients' perspectives of care and the potential effect they have on compliance, outcomes, and care-seeking behaviors. [26]

Satisfaction with care is not routinely measured in clinical research investigating neck pain conditions, despite widespread recommendations to do so. [4, 6] Previous studies similar to ours did not report patient satisfaction outcomes. [27–30] Given the disparate nature of available treatments (eg, passive vs active therapies, side effect profiles), this is a surprising and important gap.


One of the limitations of this study is that the multidimensional instrument used to measure specific satisfaction-related factors was limited to general care– and information-related domains. Currently, there is no consensus as to which satisfaction-related instruments should be used in both clinical practice and research as illustrated by the range of satisfaction measures used in neck pain research. Lack of consensus might be explained, in part, by the fact that patient satisfaction itself as a domain has been poorly researched and is still incompletely understood. [23, 26, 31, 32] Qualitative research examining the full range of issues patients consider when determining their satisfaction with care is very much needed to inform the development and choice of appropriate satisfaction measures.

Patient-provider time and attention would be expected to influence patient satisfaction but was not controlled for in this trial and should be considered a possible explanation for observed treatment group differences. However, although the average number of visits in the SMT group was 15.3 compared with 4.8 in the MED group and 2.0 for the HEA group, the similarity in many of the satisfaction-related domains between SMT and HEA suggests that time and attention had a limited affect.

The loss to follow-up, particularly at 52 weeks, was approximately 20% for the multidimensional pain instrument and highest for the MED group. Although we cannot be sure of the influence this has on study results, it could cause underestimation or overestimation on patient satisfaction. The consistency of results with other outcomes and the primary trial results limits this concern.

The exclusion of other variables for the prediction of neck pain and global improvement could be viewed as a limitation. The purpose of this analysis was to determine the contribution of specific items included in the multidimensional satisfaction questionnaire for the prediction of pain and global satisfaction, which required the exclusion of other potentially predictive variables.

Another potential limitation is that participants in the SMT and HEA groups completed the satisfaction questionnaire in the same facility where they received treatment, which may explain the higher level of reported satisfaction compared with the MED group. The observed differences in satisfaction at week 52 (all of which were collected by mail) make this unlikely and diminish this concern.


This study provides a greater understanding of satisfaction with care in acute neck pain patients receiving spinal manipulation, home exercise and advice, and medications. A consistent advantage for spinal manipulation and home exercise was identified in terms of satisfaction with general care (including provider concern, quality of treatment recommendations, and overall care) and information provided (including cause, prognosis, activities to hasten recovery, and prevention). Patients receiving spinal manipulation were also more satisfied with general care than the home exercise group. Although these secondary analyses also revealed the HEA group to be as satisfied overall with information received as the SMT group, more HEA patients were satisfied with the information received related to activities to hasten recovery and prevention. These results highlight how global satisfaction measures may mask important nuances of how patients view and experience treatments and point to the use of multidimensional satisfaction instruments in future research.


Since 9–22–2014

                  © 1995–2024 ~ The Chiropractic Resource Organization ~ All Rights Reserved