1. Top of page
  2. Abstract
  7. Acknowledgements

Clinical consequences of osteoporotic vertebral fractures, such as back pain, functional limitations, and impairment of mood, are often cited as justification for prevention and therapy. But these symptoms are poorly characterized, and a clinical grading system is not available. The aim of this study was to compare clinical measures for spinal deformation and quality of life components between patients with osteoporosis and patients with chronic low back pain (CLBP) and to determine the relationship between spinal deformation and quality of life components. A total of 130 female patients (63 osteoporotic patients, 65 ± 7.9 years, and 77 CLBP patients, 56 ± 6.5 years) had a standardized interview on quality of life components (pain, activities of daily life, mood) and clinical measures of spinal deformation (height reduction [HR], distance from occiput to wall [DOW], and distance from iliac crest to ribs [DIR]). Spinal X-rays were reviewed in all patients for the evidence of vertebral fractures. In osteoporotic patients, vertebral deformity was quantified by the spine deformity index (SDI) on X-rays. It was assessed whether subgroups could be identified by a combination of indices for spinal deformation (SDI, HR, DOW) using a cluster analysis. Back pain was a major complaint in both groups, without differences in pain intensity and frequency. Impairment of general well being and mood was found in about one-third of the patients in both groups. Independent of age, the disability score was significantly higher in patients with osteoporosis than in patients with CLBP. Both groups differed with respect to clinical measures of spinal deformity (HR, DOW, DIR). Among osteoporotic patients, parameters of quality of life were not linearly related to the degree of radiologically assessed vertebral deformity, but osteoporotic patients with two or more vertebral fractures tended to have more functional limitations than those with only one fracture. There was, however, a significant linear relationship between components of quality of life (disability score, pain) and clinical measures of spinal deformation (HR, DOW, DIR). The osteoporotic patients were subdivided into three clusters. The first group was characterized by low spinal deformation ([DOWNWARDS ARROW]SDI, [DOWNWARDS ARROW]HR, [DOWNWARDS ARROW]DOW) and little impairment of quality of life. The second group had significantly greater spinal deformation ([UPWARDS ARROW]SDI, [UPWARDS ARROW]HR, [UPWARDS ARROW]DOW) and significantly more pain and functional limitations. The third group was characterized by increased kyphosis, mainly caused by nonskeletal dysfunction ([DOWNWARDS ARROW]SDI, [DOWNWARDS ARROW]HR, [UPWARDS ARROW]DOW), but pain and functional limitations were impaired to the same degree as in the second group with severe skeletal spinal deformation. We conclude that with respect to quality of life components, functional limitation is the most specific to spinal osteoporosis and is related to clinical measures of spinal deformation. Furthermore, spinal deformation and the clinical course of osteoporosis appears to be insufficiently reflected by radiological indices of vertebral deformity (such as SDI) alone. For grading the disease and for therapeutical concepts, radiological measures and clinical evaluation should be considered in combination.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Osteoporosis is a chronic metabolic bone disease leading to progressive destruction of bone microarchitecture and loss of function. The late clinical stage of osteoporosis is characterized by insufficiency fractures which are often followed by chronic pain, physical limitations, and impaired general well being and mood. These sequelae of osteoporotic fractures are considered as major indices of an impaired quality of life(1) and are targeted increasingly by researchers, clinicians, and the pharmaceutical industry as one outcome measure for therapeutic and preventive strategies.(2-4) However, there is still no gold standard available to assess the construct of quality of life.(5) The assessment of pain, functional limitations, and emotional status represent health-related quality of life, whereas overall quality of life is a reflection of how individuals perceive their life considering also non–health-related factors such as jobs, family, or friends.

So far in clinical osteoporosis studies, outcome has exclusively been based upon hard endpoints, i.e., incident fractures. With regard to spinal osteoporosis, this is achieved by morphometric assessment of incident vertebral deformities on X-rays. Few epidemiological and clinical studies have examined the relationship between vertebral deformity and quality of life components and have generally found no or a weak association, with loss of quality of life evident only in a subgroup with severe vertebral deformation.(6-15) The relationship between spinal deformity and quality of life may be masked by indices of quality of life impairment not specific to spinal osteoporosis. In addition, the question arises as to whether quantitative assessment of vertebral deformity is sufficient to grade the clinical course of vertebral osteoporosis, because it may yield an incomplete picture of the limitations attributable to the disease. To characterize subjective and objective limitations specific to spinal osteoporosis, we compared 63 female patients with vertebral osteoporosis and 77 women with chronic low back pain (CLBP) with respect to upper body deformation (X-ray, physical examination) and components of quality of life (pain, functional capacity, mood). Within each group, we then assessed the relationship between objective measures of spinal deformity and various quality of life components. Specifically, among osteoporotics, we tested the hypothesis of whether clusters of patients could be differentiated by a combination of indices of upper body deformation, and whether these subgroups also differed with respect to functional disability and other components of quality of life.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Study population

We investigated 63 female patients with spinal osteoporosis (65 ± 7.9 years) and 77 female patients with CLBP without osteoporosis (56 ± 6.5 years). The patients with osteoporosis were admitted to the University of Heidelberg Medical Clinic, Department of Endocrinology, consecutively between February 1989 and November 1990 for further diagnosis or initiation of therapy. The patients with CLBP had been admitted to the Kraichgau-Klinik for rehabilitation and treatment of back pain. All patients were interviewed by two members of our group (C.S. and G.W.) using a standardized questionnaire and received a physical examination and X-ray evaluation of the thoracic and lumbar spine. The total interview and clinical examination lasted 1.5–2.5 h/patient.

Spinal osteoporosis was defined by the presence of at least one vertebral fracture at the thoracic or lumbar spine, provided the fracture was not attributable to a history of severe trauma, e.g., an automobile accident. Visual reading of the X-rays for the presence of osteoporotic fractures was performed at the University of Heidelberg radiology department by an experienced radiologist. Only patients who had at least one unequivocally evident vertebral fracture were included in the study; additional vertebral morphometry was performed and confirmed that vertebral height reduction was at least 20% in fractured vertebrae (see below). Secondary osteoporosis was excluded by history, laboratory tests, and within a subgroup of patients by bone histology.

None of the controls (CLBP patients) had evidence of a vertebral fracture as determined by radiological evaluation of X-rays of the thoracic and lumbar spine. Conditions underlying the back pain in CLBP patients included degenerative spine diseases like spondylosis or osteoarthritis, muscle spasm, cervical spine syndromes or reasons that could not be specified (n = 13). Several medications influencing bone metabolism (calcium, vitamin D, fluoride, calcitonin) were used by 53 women with osteoporosis over a mean duration of 1.3 ± 0.7 years and also by 15 patients with CLBP over a period of 0.3 ± 0.7 years. Estrogens were used at the time of interview by 13 (20.6%) patients with osteoporosis and by 33 (42.8%) of the patients with CLBP.


The interview is based on a composition of standardized questions or instruments regarding pain history and psychological impact.(16-22) We also included some questions from our earlier study concerning history and complaints in osteoporotic patients (i.e., circumstances when diagnosis of spinal osteoporosis was performed, functional limitations to determine a disability score).(9) The comprehensive back pain history(16-19) considered duration, frequency, pattern, localization, intensity of pain, triggers of back pain, and possibilities for coping and use of pain relievers. A list of 34 pain words (Hoppe scale, 19) to characterize pain quality was provided. Functional limitations were assessed by a disability score based on six activities of daily life and by a question about impairment of selfcare in general. Emotional status was investigated by the Center of Epidemiologic Studies depression scale (CES-D, 20) and by a general well-being scale (Zerssen Bf-S).(21,22) Details about the single items of the questionnaire are summarized in the Appendix.

Furthermore, the interview included questions regarding the personal and family medical history and known risk factors of osteoporosis (gynecological history, nutrition, exercise and mobility, medication, etc.). Data regarding the medical history were used to identify patients with secondary osteoporosis, who were excluded from this study. This respective part of the questionnaire is not further described here.

Clinical examination

Physical measurements

A comprehensive physical examination of the spine and body stature was performed. In this study the following parameters were used for further analysis.

  • • Height reduction (HR) in cm: the difference between the measured current body height and the previous body height as derived by standardized interview asking for the body height at age 25.(23,24)

  • • Determination of the distance from the iliac crest to the lowest rib (cm) (DIR) (the fingers of the investigator was interposed between the lowest ribs and the iliac crest and the respective distance was estimated with a ruler).

  • • Determination of distance from occiput to wall (cm) (DOW) in a standardized standing position (patients were instructed to stand at the wall with their back so that they touched the wall with their shoulders and were looking in a horizontal direction without forward or backward flexion of the head—the distance between the os occiput and the wall was determined with a ruler).

Functional measurements

The following functional aspects were evaluated.

  • • The ability to get up from a supine position without use of the hands (0 = possible without difficulties, 1 = possible with moderate pain, 2 = possible with severe pain, 3 = impossible).

  • • Time (in s) needed to arise from a supine position into a standing position.

Quantitative radiological assessment of vertebral deformity

In the osteoporotic patients vertebral fractures were assessed on lateral X-rays of the thoracic and lumbar spine by vertebral height measurements. We determined the spine deformity index (SDI),(25) a quantitative measurement of the extent of vertebral deformation between T5 and L5. The deformation of the spine due to fractures was determined for the whole spine (SDItotal) and for anterior vertebral deformations (SDIanterior). In addition to the SDI, we calculated the number of vertebral fractures based on the vertebral height measurements using the criteria of 20% height loss (unadjusted algorithm) as described by Melton et al.(26) We used this unadjusted algorithm for fracture definition, because in our patients diagnosis of at least one osteoporotic vertebral fracture was already confirmed by expert reading as well as by comprehensive additional diagnostic procedures, so that misclassification of vertebral deformities unrelated to osteoporosis seemed to be unlikely. Complete vertebral morphometry and determination of SDI was possible in 55 patients.

Statistical analysis

Frequency and grading of the symptoms were compared between the two patient groups by descriptive analyses using the Wilcoxon-Mann-Whitney U-test for continuous variables or ordinal scaled scores and Fisher's exact test for categorical variables. Analysis of covariance (ANCOVA) was used to assess between-group differences of age-dependent continuous variables, and direct adjustment was performed to adjust for age in categorical variables. Within groups, the correlation between different symptoms and parameters describing severity of disease were assessed by Spearman rank correlation analyses. Partial correlation analyses were performed for age-dependent variables (HR, DIR, DOW, disability score).

Among patients with osteoporosis, a cluster analysis was performed to assess whether different subgroups of patients could be identified by a similar combination of three parameters describing spinal deformation (SDI—radiological assessment) and total upper body deformation (HR, DOW—clinical measures). Using this combination of variables (SDI, HR, DOW) we could identify three subgroups of osteoporotic patients. Differences between the three groups in the parameters describing spinal deformation as well as in quality of life components were assessed by analysis of variance (ANOVA) and Scheffe's test. Because of multiple tests (comparisons between the two groups and correlations being evaluated within the groups) we performed Bonferroni's adjustment and set the level of significance to p = 0.001. All procedures were performed with statistical analysis system, SAS 6.04 (SAS Institute, Cary, NC, U.S.A.).


  1. Top of page
  2. Abstract
  7. Acknowledgements

Comparison between the osteoporotic and CLBP patient groups

The distribution of age, parameters of clinical investigation, and variables of quality of life among patients with osteoporosis and CLBP is shown in Table 1. Age, disability score, self-care in general, and the parameters of clinical investigation (HR, DOW, DIR) were significantly different between both groups; no differences were found between the sum of pain words, general well-being scale, and depression scale.

Table Table 1. Distribution of Age, Components of Quality of Life (Pain, Functional Limitations, and General Well Being and Depression Scale) and Parameters of Clinical Investigation Among Osteoporotic (n = 63) and Chronic Low Back Pain Patients (CLBP, n = 77) CLBP
Thumbnail image of

If both patient groups were assessed together, we found a significant correlation between age and the following variables: HR, r = 0.6, p < 0.0001; DOW, r = 0.44, p < 0.0001; DIR, r = −0.45, p < 0.0001; disability score, r = 0.25, p < 0.003, and self care in general r = 0.27, p < 0.001. Subsequent analyses including these age-dependent variables were controlled for age influence. The parameters of pain history (pain intensity, frequency, and sum of pain words), general well being and mood (Bf-S Zerssen, CES-D) were independent from age as well as SDI within the osteoporotic patients.

Functional limitations

The frequency of limitations in walking, bending, dressing, carrying bags, climbing stairs, and rising from a lying position is shown for both patient groups in Table 2. The disability score was significantly higher in patients with osteoporosis (mean 4.6 ± 2.8, median 4.5) than in patients with CLBP (mean 2.6 ± 2.2, median 2) (p < 0.0001). The differences between the groups persisted after adjustment for age by ANCOVA (age-adjusted mean ± SD, 4.3 ± 2.6 and 2.8 ± 2.5; p < 0.002 in osteoporotic vs. CLBP patients, respectively). All individual components of the disability score were rated significantly higher by the osteoporotic than by the control group, but carrying bags and bending showed the most pronounced differences. The age-adjusted rate of patients with limitations in self-care in general was 84% in osteoporotic patients (in 36% self-care was more difficult, but possible alone; 48% were dependent on assistance in some aspects of self-care) as opposed to 63% in CLBP patients (in 54% self-care was more difficult, but possible alone; 9% were dependent on assistance in some aspects of self-care). Patients with osteoporosis suffered from a significantly stronger limitation in self-care in general than patients with CLBP (p < 0.0001).

Table Table 2. Frequency of Limitations During Daily Life Activities in Osteoporotic Patients (N = 61) and Chronic Low Back Pain (CLBP, N = 77) Patients
Thumbnail image of
Back pain history

Among patients with vertebral osteoporosis, the mean duration of back pain and complaints leading to a diagnosis of osteoporosis was 4 ± 4.9 years. However, the overall duration of back pain the patients had experienced before the interview was 12.9 ± 11.4 years in the patients with osteoporosis and 13.7 ± 9.3 years in those with CLBP. Seventy-three percent of our patients with osteoporosis reported a sudden onset of pain, as compared with 21% of the patients with CLBP (p < 0.001). In a more detailed question on circumstances when diagnosis of osteoporosis was performed, 13% of patients with osteoporosis remembered an accident at home (fall, stumble), 24% remembered having lifted a heavy load, and 44% had a sudden onset of pain without any evident reason, whereas 19% reported a more gradual begin of complaints. Severe back pain (71%) and height reduction (10%) were the main events that had triggered further diagnostic procedures in the patients with osteoporosis, before their diagnosis was established. In 21%, the diagnosis of osteoporosis was made incidentally, when consulting a physician for other reasons. The median number of physicians consulted due to back pain was three in both patient groups and ranged from one to seven. In both patient groups, general practitioners and specialists in orthopedics and medicine were the most frequently consulted physicians. More than half of the patients with osteoporosis (55%) and also of the patients with CLBP (62%) had stayed in a hospital or rehabilitation center for a mean duration of 4 weeks due to back pain. In addition to the hospital stays, nearly all patients (84% of the patients with osteoporosis and 97% of the patients with CLBP) had been treated with at least one course of physicotherapy (e.g., massage, heat application, or remedical gymnastic). At the time of the interview, 30% of patients with osteoporosis and 12% of patients with CLBP reported the current use of analgesics.

In both patient groups, pain was mainly located in the lumbar spine (osteoporosis 80%, CLBP 89%). However, osteoporotic patients experienced pain in the thoracic spine more frequently than CLBP (osteoporosis 64%, CLBP 38%), whereas the opposite was true for pain located at the cervical spine (osteoporosis 31%, CLBP 72%). The dependency of pain on physical activity was more pronounced in patients with osteoporosis than in those with CLBP. Pain was only present in combination with physical activity in 56% of the patients with osteoporosis as compared with 29% in patients with CLBP (p < 0.002). More than 70% of the patients from both groups graded their pain at the time of interview as being mild or moderate, but more than 90% of the patients of either group graded the worst back pain they had ever experienced as severe or intolerable. We found no circadian pattern of pain in either patient group. There was no significant difference in the frequency of back pain at all and the frequency of night pain between the two patient groups. In either group, about 60% of the patients suffered from daily pain and 40% of the patients experienced night pain at least once per week. In both patient groups, 30% of the patients could not relate their pain to bone or muscle nor distinguish between superficial or deep pain. The majority of patients in either group perceived their pain as being deep (osteoporosis 59%/CLBP 70%) and attributed pain to bone (osteoporosis 42%/CLBP 41%).

Table 3 shows the five most frequently stated ways to improve or worsen pain in osteoporotic patients and the frequency of the respective statements in CLBP patients. Resting, lying down, physicotherapy, heat application, and moving around were the most frequently stated methods to relieve pain in osteoporotic patients, whereas remaining in a given position, bending, standing, rising from a lying position, and doing housework caused aggravation of pain. Lying down was more frequently stated in osteoporotic patients to relieve pain than in CLBP patients (p < 0.02). Complementary physical activity like doing housework was more often stated to worsen pain in osteoporotic patients than in CLBP patients (p < 0.008). Beyond this it was remarkable that massage relieved pain in 86% of patients with CLBP but only in 49% of patients with osteoporosis and was considered even to worsen the pain in 27% of the patients with osteoporosis in comparison with 7% of the patients with CLBP.

Table Table 3. Parameters Influencing Back Pain in Osteoporotic Patients (N = 61) and Chronic Low Back Pain (CLBP) Patients (N = 77)
Thumbnail image of
Emotional status

The means of the depression score (CES-D) and general well being scale (Bf-S Zerssen) were not different between both patient groups (Table 1). The Zerssen scale was in the normal range in 75% of the osteoporotic and in 74% of the CLBP patients. According to the Zerssen scale, clearly limited well being (score > 27) was found in 12% of the osteoporotic and in 11% of the CLBP patients, whereas 13% of the osteoporotic and 15% of the CLBP patients were considered to have borderline abnormal well being (score 22–27). According to the CES-D depression scale in both groups, a remarkable high proportion was found to have values above 16, indicating depressive symptoms (40% of the osteoporotic patients and 35% of the CLBP patients).

There was a consistent positive relationship between pain, as assessed by the Hoppe scale, and the general well-being scale in both patient groups, with correlation coefficients of r = 0.57 (p < 0.0001) in patients with osteoporosis and r = 0.72 (p < 0.0001) in patients with CLBP. The correlation coefficient between the general well-being scale (Bf-S Zerssen) and the scale for depressive symptoms (CES-D) was r = 0.67 (p < 0.0001) in patients with osteoporosis and r = 0.69 (p < 0.0001) in patients with CLBP. Furthermore, we found strong positive correlations between pain (Hoppe scale), general well-being scale (Bf-S Zerssen), and depression scale (CES-D) on the one hand and the disability score on the other hand in both patient groups (Table 4).

Table Table 4. Spearman Partial Correlation Coefficients (Controlling for the Effect of Age) of Parameters of Clinical Investigation and Quality of Life Components (Sum of Pain Words, General Well-Being Scale and Depression Scale) Versus Disability Score (0–12) in Osteoporotic Patients (N = 60) and Chronic Low Back Pain Patients (CLBP, N = 74)
Thumbnail image of

Relationship between spinal and upper body deformation and components of quality of life

In osteoporotic patients, the degree of vertebral deformities was quantified on X-rays by SDI. The mean value for SDItotal was 1.9 ± 1.65 T4 units, median 1.35, and for SDIanterior 0.84 ± 0.71 T4 units, median 0.7. The mean number of vertebral fractures was 3.8 ± 2.5. The osteoporotic patients were stratified into subgroups with one vertebral fracture (n = 10) and two or more vertebral fractures (n = 45).

The disability score of daily life activities was significantly correlated with all three clinical measures of spinal deformity (HR, DOW, DIR) in osteoporotic patients, but not in CLBP patients (Table 4). The group of osteoporotic patients who were limited within their independence in self-care had a significantly higher deformation of body shape (HR, DOW, DIR) than patients without limitations (Fig. 1). These differences persisted after correction for age dependency of HR (p < 0.01), DOW (p < 0.001), and DIR (p < 0.006). We found no such differences in the patients with CLBP. SDI indices were neither significantly correlated to the disability score nor significantly different in osteoporotic patients with and without limitations regarding basic activities of daily life. The number of pain words (Hoppe scale) significantly correlated with parameters of clinical investigation (HR, r = 0.37, p < 0.004; DOW, r = 0.46, p < 0.0003; DIR, r = −0.41, p < 0.001) in patients with osteoporosis but not to SDItotal or SDIanterior.

thumbnail image

Figure FIG. 1. Relationship between self-care in general and height reduction (bottom), distance from occiput to wall (middle), and distance from iliac crest to ribs (top) in patients with spinal osteoporosis. Self-care in general was rated into three categories: possible without difficulties (= group A, left); possible with difficulties and increased time (= group B, middle) and impossible, being dependent on extra help (= group C, right). ▴ Significant differences between patients without difficulties (group A, left) and those being dependent on extra help (group C, right) (HR, p < 0.015; DOW, p < 0.001; DIR, p < 0.006). ▴▴ Significant difference between patients without difficulties (group A, left) and patients with difficulties (group B, middle) (DOW, p < 0.04).

Download figure to PowerPoint

In patients with osteoporosis, the score of the general well-being scale (Bf-S Zerssen) was correlated with DOW (r = 0.32, p < 0.001) and DIR (r = −0.41, p < 0.001) but not with HR and SDI. Depressive symptoms (CES-D scale) were not correlated to parameters of physical examination in either patient groups nor with SDI in patients with osteoporosis. In addition, we compared the clinical outcome within the group of osteoporotic patients after stratification into those with only one vertebral fracture and those with two or more fractures. Table 5 shows the distribution of health related quality of life components and the parameters of clinical investigation (HR, DOW, DIR) among these two subgroups of osteoporotic patients. Patients with two or more vertebral fractures had significantly more spinal deformation determined by clinical measures (HR, DOW, DIR) than those with only one vertebral fracture. We found a clear tendency that patients with two or more vertebral fractures suffered from more severe functional limitations during activities of daily life than those patients with only one vertebral fracture (p < 0.025). The other parameters like pain and mood showed a similar trend.

Table Table 5. Distribution of Age, Parameters of Clinical Investigation, and Major Variables of Quality of Life Within Osteoporotic Patients After Stratification into a Group with Only One Vertebral Fracture (n = 10) and Two or More Vertebral Fractures (n = 45)
Thumbnail image of

Subclassification of patients with vertebral osteoporosis

A cluster analysis was performed to identify subgroups of patients with osteoporosis based on a combination of variables representing spinal and upper body deformation (SDI, HR, DOW). We could thereby identify three clusters of osteoporotic patients. Table 6 gives mean values and standard deviations for SDI, HR, and DOW within the three clusters of osteoporotic patients. Figure 2 shows the distribution of SDI, HR, and DOW for all patients within the three clusters. The distribution of quality of life components (disability score, self-care in general, general well-being scale, depression scale, and pain–Hoppe scale) and age are summarized in Table 6 for the three subgroups of osteoporotic patients.

Table Table 6. Distribution (Mean ± SD) of Height Reduction, Distance from Occiput to Wall and Spine Deformity Index (T4 Units) as Well as Age and Quality of Life Components Within Three Clusters of Patients With Osteoporosis (n = 52)
Thumbnail image of
thumbnail image

Figure FIG. 2. Subclassification of patients with osteoporosis into three subgroups by a cluster analysis using the variables spine deformity index (T4 units), height reduction (cm), and distance from occiput to wall (cm); (Cluster 1 inline image – SDI [DOWNWARDS ARROW], HR [DOWNWARDS ARROW], DOW [DOWNWARDS ARROW]; Cluster 2 inline image – SDI [UPWARDS ARROW], HR [UPWARDS ARROW], DOW [UPWARDS ARROW]; Cluster 3 inline image – SDI [DOWNWARDS ARROW], HR [DOWNWARDS ARROW], DOW [UPWARDS ARROW]).

Download figure to PowerPoint

Patients in cluster one were characterized by relatively low SDI values, low DOW, and little HR. Functional limitations and pain were also less in this group and significantly lower in comparison with patients of cluster two. These patients were characterized by significantly higher values of SDI, HR, and DOW compared with cluster one. Patients of cluster three differed from cluster one only with respect to a significantly higher DOW but not in SDI and HR. However, functional limitations and pain were significantly higher in group three compared with group one and did not differ from cluster two, which was characterized by a high level of spinal deformation determined by radiological as well as clinical criteria. Patients of cluster two and three had significantly more frequent limitations in self-care in general than patients of cluster one. Differences in age, in the well-being scale, and depression scale between the three clusters did not reach statistical significance.


  1. Top of page
  2. Abstract
  7. Acknowledgements

The results of this study demonstrate that clinical symptoms and complaints are not specific for osteoporosis because similar findings were noted in CLBP patients. Functional limitations were merely more pronounced in osteoporotic patients than in CLBP patients. Pain was a major complaint in nearly all patients with osteoporosis, but there was no specific pain pattern. In both osteoporotic and CLBP patients, impaired well-being due to back pain was found without differences in the severity or frequency of the latter. This indicates that the underlying cause of the back pain was not a major determining factor for the subjective outcome. Clinical measures indicating spinal deformation (HR, DOW, and DIR) differed between the two groups: all three parameters were significantly more pronounced (HR [UPWARDS ARROW], DOW [UPWARDS ARROW], DIR [DOWNWARDS ARROW]) in the osteoporotic group compared with the control group. In osteoporotic patients, these clinical measures describing overall spinal deformation (skeletal and nonskeletal causes) were associated with the health-related quality of life components. In contrast, the radiological assessment of SDI showed no such association. This observation emphasizes the importance of nonskeletal parameters for the grading of subjective sequelae associated with vertebral osteoporotic fractures.

Our study population consisted of: (1) patients consecutively admitted into the Heidelberg University Hospital with either a previously established diagnosis of osteoporosis or suspected diagnosis of osteoporosis, and (2) patients with CLBP who were admitted into a rehabilitation clinic. We cannot exclude diagnostic bias due to lack of blinding of the interviewers with respect to the diagnosis of the patients. However, differences between CLBP and osteoporotic patients were limited to functional disabilities and limitation in self-care in general and there was no difference regarding pain and mood, which could have been subject of bias as well. Beyond this it remains uncertain whether the patients of our study are actually representative of all osteoporotic and all CLBP patients. Therefore, these results should be confirmed by future studies within a representative survey. Because it may be difficult for medical staff to oversee clinical signs of severe osteoporosis, interview performance by nonmedical trained persons may help to reduce the possibility of interviewer bias.

Our questionnaire was developed to assess quality of life components in patients with vertebral osteoporotic fractures and patients with CLBP in a standardized way. Quality of life is difficult to assess because it is highly subjective and variable as a concept.(1,5,27) Because functional abilities, pain and mood are widely accepted as major components of quality of life, our questionnaire focused upon these. The questionnaire was developed out of different sections from several pain questionnaires and psychometric scales, which had previously been standardized and validated.(16-22) Visual analog scales (VASs) are widely used to assess intensity of pain. After a test phase, we decided not to use VAS in our questionnaire, since many of the older people had problems understanding and applying this kind of representation. The retest and interrater reproducibility of the entire questionnaire could not be determined because patients could not be expected to answer a lengthy questionnaire (1.5–2.5 h) for a second time.

Considering the recall of body height at the age of 25 years to calculate HR is suboptimal, because historical data on body height are probably inaccurate. However, we felt it important to keep HR within our analyses for several reasons. First, HR is generally accepted as a clinical feature indicating manifest spinal osteoporosis and is recommended for the evaluation of patients with suspected osteoporosis in clinical practice although measurements of previous body height obtained under standardized conditions are rarely ever available. Second, the moderate but significant correlation between HR and SDI an objective radiological assessment of vertebral deformity (r = 0.49) underlines the validity of HR as a clinical measure for spinal osteoporosis. Third, recall bias is probably the same in both groups. Furthermore, recall of previous body height as performed within our study has proven to be highly reproducible(23) and has been used within epidemiological surveys focusing on the prevalence and incidence of spinal osteoporosis.(23,24)

It is one limitation of our study that the questionnaire considered only the health-related aspects of quality of life and did not include a section or single question about overall quality of life. Overall quality of life can be markedly influenced by nonmedical aspects and may be graded differently between persons who have similar ratings among the health-related components. Therefore, our results do not allow a conclusion as to which extent the measured components, such as functional disabilities, pain, and mood actually influenced the overall quality of life in our patients. However, because the sequelae of vertebral fractures is not well known at all, it was the primary aim of this study to assess whether osteoporotic-specific items of health-related quality of life can be identified that should be included in future studies.

The results regarding the comparison of pain and functional limitations between the osteoporotic versus the CLBP groups could have been confounded by age because patients with osteoporosis were an average of 9 years older than patients with CLBP and because functional disability was positively correlated to age. However, adjustment for age did not explain the increased functional limitations within the osteoporosis group. However, no influence of age on general well being and pain was observed in this study. This is consistent with results from previous studies,(6,28-30) which found either no relationship between age and pain or even a tendency that pain prevalence decreased with advancing age after 65 years. Taken together, these data indicate that within the age range 50–79 years, impairment of general well being with respect to pain is at least not positively associated with age.

A comparison of the pain history between both patient groups did not reveal a pain pattern specific of osteoporotic vertebral fractures. This may be, at least in part, due to the fact that patients in the two groups suffer from chronic pain, which is known to be perceived disconnected from the actual somatic lesion. Furthermore, pain resulting from extraskeletal structures is probably relevant in both patient groups, and patients with osteoporotic fractures may in addition suffer from some conditions causing pain in the CLBP group. Several cross-sectional back pain studies(29,31-34) have shown that 41–67% of the adult population have suffered at least one episode of back pain, but only in 20% was the underlying cause ever identified. Nevertheless, most of our osteoporotic patients were able to remember the time of onset of pain and complaints that they related to osteoporosis distinctly from previously experienced pain and back problems. The duration of symptoms related to osteoporosis was remarkably shorter (4 ± 4.9 years) than the overall duration of back problems (12.9 ± 11.4 years). In contrast to the previously held view that vertebral fractures due to osteoporosis are generally atraumatic, there is increasing evidence that minor trauma, load, or other physical effort play an important role for vertebral fracturing in osteoporosis.(35) Our observation that 73% of the patients with osteoporosis reported an acute onset of pain or symptoms and that more than half of them (66%) were able to recall a specific set of circumstances as the possible cause of pain or fracturing is in accordance with this concept and with results from Cooper et al.(35) This underlines the recommendation that the clinical evaluation of patients with osteoporosis should include concrete questions about minor trauma or physical activities, especially because patients often do not spontaneously report their complaints and associated circumstances.(12)

A variety of items was stated in both groups to relieve pain or to aggravate pain (Table 3), suggesting that coping mechanisms are highly individual. A remarkable difference between the two patient groups was that massage was experienced more often as positive in CLBP patients (86%) than in osteoporotic patients (49%); in fact, 27% of patients with osteoporosis experienced worsening of their back pain with massage. This makes intuitive sense since massage adds load to fragile skeletal tissue and may even risk progressive fracturing of the vertebrae. Therefore, massage therapy in patients with osteoporosis should be done restrictively.

As far as functional limitations, difficulties with carrying bags and bending were the most frequently reported items, suggesting that these limitations may be typical for spinal osteoporosis. Cook et al.(12) showed in a comparable group of osteoporotic patients that difficulties in carrying and lifting as well as doing housework were among the most important issues evaluated by the patients to influence their quality of life. Complementary to the disability score which considers single activities, the rating of “self-care in general” is more comprehensive, indicating capability of independency at home. A remarkably higher proportion of osteoporotic patients than patients with CLBP was dependent on extra help for some aspects of self-care. This difference may be due to the specific limitations experienced by patients with osteoporosis (e.g., carrying bags). However, dependency on extra help during self care in general may reflect a coping strategy of patients who avoid certain activities triggering pain. Gold et al.(36) first addressed that the effective use of coping strategies may significantly improve the subjective health status and quality of life in patients with osteoporosis.

Impairment of psychological well being and suffering due to back pain is known from several investigations,(19,37-43) although the causal chain between pain and depression is not well understood and rates about depression in chronic pain patients vary remarkably. According to the Zerssen well-being scale, we found obvious impairment of well being in about 10% of the patients in both groups. In contrast, there was a remarkably higher proportion classified as depressive by the CES-D scale (35–40%) in both groups. This difference and the relatively high rate of depressive symptoms could be partly caused by methodological shortcomings in the assessment of depression, because insufficient specificity is one known problem of the CES-D scale.(44) The correlation between the CES-D scale and the Zerssen scale was relatively high in both groups (r = 0.67), but obviously both scales must measure at least some different aspects. Beside these methodological aspects, it has to be considered that in both groups an increased rate of depressive symptoms is actually present. Similar to our results, Cook et al.(12) also showed in their study that emotional dysfunction is one serious problem in patients with spinal osteoporosis because a considerable proportion of their patients (>50%) suffered from emotional dysfunction which was associated with functional disabilities. In both groups, we found a strong, positive correlation between the disability score and emotional status (CES-D, BF-S-Zerssen), which could be explained in several ways: (1) limitations in daily life activities and self-care in general influence overall well being and lead to a deterioration of emotional status; (2) depression is associated with reduced mobility independent from the underlying cause; (3) there may be no direct relationship between depression and functional limitations, rather the association may be spurious, resulting from the influence of chronic pain on both constructs.

The influence of bone-related drug treatment on quality of life, especially of estrogens and calcitonin, could have affected the respective ratings in both groups. Whereas calcitonin was currently used at the time of the interview only in a minority of patients in both groups (<10% of the patients), the use of estrogens could have biased the results of our study on quality of life issues. But the major components of quality of life (pain, mood, functional disability) as well as HR, DOW, DIR, and SDI were not different between patients currently using estrogens or not in both patient groups (data not shown); the only difference was that osteoporotic patients using estrogens were significantly younger than those who did not. It therefore seems unlikely that differences in medications did bias the results of this study. The limited number of our patients did not allow us to stratify the analyses by therapy regimens. To clarify the role of medication on quality of life should be one goal of clinical trials in the future.

Relationship between components of quality of life and spinal deformation

Skeletal spinal deformation was assessed by the number of vertebral fractures and by the quantitative measure of SDI. Because the clinical significance of vertebral fractures is still often questioned, we compared osteoporotic patients with only one fracture to those with advanced disease with two or more fractures. Patients with two or more vertebral fractures tended to suffer from more functional limitations and pain than those with only one vertebral fracture (Table 5). This supports the importance of consequent treatment in patients with mild manifest osteoporosis (only one vertebral fracture) and in those without fractures but low bone mass because these patients are known to have an increased risk of vertebral fractures.

But neither the disability score, the limitations in self-care in general, nor the parameters of pain history and emotional status were significantly correlated to the quantitative radiological measurement of SDI. This implies that the objectively assessed degree of skeletal spinal deformity is not necessarily related to subjective well being and function, as was also shown by others.(12) The lack of a significant linear association between SDI and functional limitations appears to be in contrast with our previous observations in patients with advanced spinal osteoporosis.(9) In the present study, there were relatively few patients with severe skeletal deformity (SDI values > 3), which may explain the lack of an association to functional disability observed now. However, the previous study mainly included patients with advanced vertebral deformities and the relationship between SDI and symptoms was evident due to those patients with high SDI values, whereas there was a wide variability in clinical symptoms in patients with low SDI values. This is in accordance with the findings of Ross et al.(10) and Ettinger et al.,(11) who found a relationship between vertebral fractures and pain/functional limitations only among patients with severe vertebral fractures. Furthermore, it is generally assumed that impairment of well being, functional disability, pain, and the deformation of body shape in patients with osteoporosis is not only related to measurable fractures but may also be influenced by the presence of microfractures, imbalance and tension of muscles, irritation of joints and ligaments, and the status of intervertebral disks.(11,14,45)

The importance of nonskeletal parameters in spinal osteoporosis became evident in our study by the strong association between clinical measures describing overall upper body deformation (HR, DOW, DIR) and quality of life components (pain, disability, emotional status) but by the absence of such a relationship between the radiological measure of SDI and quality of life components. Similar results have been addressed already by Ross et al.(10) who found height reduction to be a better determinant of the degree of limitations than a radiologically based fracture score. This means that a similar degree of skeletal spinal deformity may still result in different body shapes, depending on the patient's ability to counter the skeletal deformity by the use of muscles and ligaments. This mechanism is supported by data of Sinaki et al.,(45) who found reduced muscle strength in the back muscles of patients with osteoporosis. De Smet et al.(46) reported that although Cobbs ankle, a measure of kyphosis, was highly correlated with ventral fracturing, a significant kyphosis was also found in patients without any vertebral height deformation or fracturing. In addition, the importance of muscle strength for posture is underlined by a recently performed study of Itoi et al.(47) showing that increasing the back extensor strength by exercise in healthy estrogen-deficient women helps to decrease thoracic kyphosis.

Furthermore, the results of the cluster analysis support the importance of the nonskeletal parameters of spinal deformation in patients with manifest osteoporosis. Besides one subgroup with low spinal deformation ([DOWNWARDS ARROW]SDI, [DOWNWARDS ARROW]DOW, [DOWNWARDS ARROW]HR) which was associated with relatively few clinical symptoms, and a second group with severe spinal deformation ([UPWARDS ARROW]SDI, [UPWARDS ARROW]DOW, [UPWARDS ARROW]HR) associated with severe clinical symptoms, we identified a third group characterized by low skeletal deformation in combination with a severe kyphosis ([DOWNWARDS ARROW]SDI, [UPWARDS ARROW]DOW, [DOWNWARDS ARROW]HR) and severe clinical limitations not different from the extent of complaints in subgroup two. One explanation for the symptoms in this third subgroup is probably the increased kyphosis of the spine due to nonskeletal insufficiencies (i.e., muscle weakness, disc damages). Within a recent population-based study, Ettinger et al.(14) found no influence of the degree of kyphosis on pain and disability, but their results are not directly comparable to our findings because their study population is markedly different, including more persons with no limitations and no fractures at all and fewer persons with severe disabilities and severe fractures. Another explanation for the symptoms within the third subgroup could be the time of fracturing: recent fracturing, even if measurable skeletal deformation is low, is probably associated with more or severe symptoms than is a similar degree of fracturing which occurred years before. Because skeletal deformation in this third subgroup is relatively low and kyphosis is probably caused by muscle weakness and tension due to pain, there is a good chance that these patients will benefit particularly from intensive physical therapy. Taken together, these data support the hypothesis that skeletal deformity does not explain the grade of the clinical presentation of vertebral osteoporosis alone.

We conclude that for a grading of spinal osteoporosis, both clinical and radiological measures for spinal deformity should be considered. Because there is no simple relationship between measures of spinal deformation and subjective health status and limitations, it is also not possible to leave quality of life components such as pain, functional disabilities, and general well being aside, exclusively relying on parameters of clinical or radiological examination. This implies that determination of outcome measures in clinical trials in patients with osteoporosis should be broadened and bone-related measurements (bone density, X-ray evaluation, and laboratory findings) should not be the only endpoints in evaluating therapeutic effects. Furthermore, the importance of nonskeletal parameters with respect to subjective health status underlines the need to assess the efficacy of therapeutic strategies other than medication like physical therapy, pain therapy, and coping strategies in a systematic way.


  1. Top of page
  2. Abstract
  7. Acknowledgements

We are grateful to our patients who participated in this study. We also would like to thank Dr. H. Seemann, Clinic of Psychosomatic, University of Heidelberg, for her development of the questionnaire.


  1. Top of page
  2. Abstract
  7. Acknowledgements
Pain history(16-19)
  • • Time between first back complaints and diagnosis (in years).

  • • Reasons for initiation of diagnostic procedures (pain, height reduction, fractures, incidental diagnosis).

  • • Circumstances concerning onset of osteoporosis-related back pain (carrying heavy loads, fall, sudden movement).

  • • Onset of first back pain (sudden, gradual).

  • • Duration of back pain (in years).

  • • Frequency of pain: never, once or several times/month, once or several times/week, daily.

  • • Pain dependent on: (1) time of day, (2) physical activity.

  • • Pattern of pain: (1) continuous, steady, constant, (2) rhythmic, periodic, intermittent, (3) brief, momentary, transient.

  • • Perception of pain as: (1) muscle pain and/or bone pain, (2) superficial and/or deep.

  • • Localization of pain: cervical spine, thoracic spine, lumbar spine, elsewhere.

  • • Frequency of night pain: never, once or several times/month, once or several times/week, daily.

  • • Sleep disturbances: difficulties in falling asleep, awaking due to back pain.

  • • Intensity of pain: none, mild, moderate, severe, intolerable (0–4).

  • • Worst back pain intensity ever experienced in comparison to worst experienced abdominal pain, headache, and toothache.

  • • Number of different physicians consulted due to back pain.

  • • Number of courses of physical therapy.

  • • Number of hospitalizations due to back pain.

  • • Use of pain relievers.

  • • Triggers of back pain and possibilities for coping were assessed by a list of 20 items (physical, environmental, and psychological factors) to describe their influence on back pain (no influence, relief of pain, aggravation of pain).

  • • Hoppe scale(19) (list of 34 pain words). The pain words can be grouped into a sensory and affective category, which are summed to one score. This sum score out of the number of pain words is also used as a indirect measure of pain intensity.

Functional limitations

A sum score for functional disabilities (range 0–12) was calculated based on the rating of six activities of daily life (walking, bending, dressing, carrying bags, climbing stairs, getting up from lying positions [0 = possible without difficulties, 1 = possible with difficulties, 2 = dependent on help or impossible]). Cronbach's ∝ for this disability score was 0.87 if both groups were assessed together; it was not different between both groups if determined separately. Self-care in general was assessed as follows: 0 = possible without difficulties, 1 = possible with difficulties and increased time, 2 = dependent on extra help, 3 = nursing care needed.

Impairment of mood
Center of Epidemiologic Studies depression scale (CES-D)(20)

It consists of 20 questions including cognitive, affective, behavioral, somatic symptoms, and positive affect to assess depressive symptomology. Response categories range from “rarely or none” to “most or all of the time” and are based on a stated time range of less than 1 day to 5–7 days of the previous week (0 = never; 1 = sometimes, 1–2 days a week; 2 = frequent, 3–4 days per week, and 3 = 5–7 days per week). The sum score ranges from 0 to 60, and a cutting point of 16 is commonly used to identify subjects with depressive symptoms.

General well being was assessed by Zerssen's scale (Bf-S)(21),(22)

The Zerssen scale gives a quantitative estimate for the impairment of subjective well being by general complaints (e.g., fatigue, weakness) and localized somatic symptoms (e.g., cardiovascular, neuromuscular complaints). The addressed 24 different items range from mainly somatic symptoms (e.g., loss of appetite, heartburn, or articular pain) to mainly psychic complaints (anxiety, rumination) and includes symptoms between these two categories (e.g., loss of sleep, weakness of concentration). Each item was graded from 0 to 3 (0 = absent, 1 = mild, 2 = moderate, 3 = severe). The scores may range from 0 to 72, with a score of 0–21 considered normal, 21–27 as probably disturbed well being, and a score of >28 as definitely disturbed well being. This scale was applied as an additional measure for general well-being and mood impairment because it has been developed and validated within the German speaking area and has been shown to have good reliability; normal values from a representative population were available (mean 14.3 ± 10.7), and the score is independent of age or social class.

  • 1
    Schipper H, Levitt M 1985 Measuring quality of life: Risks and benefits Cancer Treat Rep 69: 11151125.
  • 2
    Kanis JA, Minne WH, Meunier PJ, Ziegler R, Allender E 1992 Quality of life and vertebral osteoporosis (editorial) Osteoporosis Int 2: 161163.
  • 3
    Kleerekoper M, Nelson DA, Peterson EL, Tilley BC 1992 Outcome variables in osteoporosis trials Bone 13: 2934.
  • 4
    Greendale GA, Silverman SL, Hays RD, Cooper C, Spector T, Kiel D, Reuben DB 1993 The Osteoporosis Quality of Life Study Group: Health-related quality of life in osteoporosis clinical trials Calcif Tissue Int 53: 7577.
  • 5
    Gill TM, Feinstein AR 1994 A critical appraisal of the quality of quality-of-life measurements JAMA 272: 619626.
  • 6
    Finsen V 1988 Osteoporosis and back pain among the elderly Acta Med Scand 223: 443449.
  • 7
    Bergenudd H, Nilsson B, Udén A, Willner S 1989 Bone mineral content, gender, body posture and build in relation to back pain in middle age Spine 14: 577579.
  • 8
    Ettinger B, Block JE, Smith R, Cummings SR, Harris ST, Genant HK 1988 An examination of the association between vertebral deformities, physical disabilities, and psychosocial problems Maturitas 10: 283296.
  • 9
    Leidig G, Minne HW, Sauer P, Wüster C, Wüster J, Lojen M, Raue F, Ziegler R 1990 A study of complaints and their relation to vertebral destruction in patients with osteoporosis Bone Miner 8: 217229.
  • 10
    Ross PD, Ettinger B, Davis JW, Melton LJ, Wasnich RD 1991 Evaluation of adverse health outcomes associated with vertebral fractures Osteoporosis Int 1: 134140.
  • 11
    Ettinger B, Black DM, Nevitt MC, Rundle AC, Cauley JA, Cummings SR, Genant HK 1992 Contribution of vertebral deformities to chronic back pain and disability J Bone Miner Res 7: 449456.
  • 12
    Cook DJ, Guyatt GH, Adachi JD, Clifton J, Griffith LE, Epstein RS, Juniper EF 1993 Quality of life issues in women with vertebral fractures due to osteoporosis Arth Rheum 36: 750756.
  • 13
    Lyles KW, Gold DT, Shipp KM, Pieper CF, Martinez S, Muhlhausen PL 1993 Association of osteoporotic vertebral fractures with impaired functional status Am J Med 94: 595601.
  • 14
    Ettinger B, Black DM, Palermo L, Nevitt MC, Melnikoff S, Cummings SR 1994 Kyphosis in older women and its relation to back pain, disability and osteopenia: The study of osteoporotic fractures Osteoporosis Int 4: 5560.
  • 15
    Greendale GA, Barrett-Connor E, Ingles S, Haile R 1995 Late physical and functional effects of osteoporotic fractures in women: The Rancho Bernardo Study J Am Geriatr Soc 43: 955961.
  • 16
    Seemann H 1987 Anamnesen und Verlaufsprotokolle chronischer Schmerzen für die Praxis - Ein Überblick Der Schmerz. Springer Verlag, Berlin, Germany 1: 312.
  • 17
    Melzack R, Torgerson WS 1971 On the language of pain Anaesthesiology 34: 5059.
  • 18
    Melzack R 1975 The McGill Pain questionnaire: Major properties and scoring methods Pain 1: 227299.
  • 19
    Hoppe F 1985 Zur Faktorenstruktur von Schmerzerleben und Schmerzverhalten bei chronischen Schmerzpatienten Diagnostica 31: 7078.
  • 20
    Radloff LS 1977 The CES-D scale: A self-report depression scale for research in the general population Appl Psychol Meas 1: 385401.
  • 21
    Zerssen D 1976 Die Beschwerdenliste/Manual unter Mitarbeit von D.M. Koeller aus dem Münchner Psychiatrischen Informationssystem (Psychis, München), Beltz Verlag, Weinheim, Germany.
  • 22
    Zerssen D, Koeller DM, Rey ER 1970 Die Befindlichkeitsskala (Bf-S) ein einfaches Instrument zur Objektivierung von Befindlichkeitsstörungen, insbesondere im Rahmen von Längsschnittuntersuchungen Arzneimittel Forsch 20: 915918.
  • 23
    O'Neill TW, Cooper C, Cannata JB, Lopez JBD, Hoszowski K, Johnell O, Lorenc RS, Nilsson B, Raspe H, Stewart O, Silman AJ, on behalf of the European Vertebral Osteoporosis Study (EVOS) Group 1994 Reproducibility of a questionnaire on risk factors for osteoporosis in a multicentre prevalence survey: The European Vertebral Osteoporosis Study Int J Epidemiol 23: 559564.
  • 24
    Black DM, Palermo L, Nevitt MC, Genant HK, Epstein R, San Valentin R, Cummings SR 1995 Comparison of methods for defining prevalent vertebral deformities: The study of osteoporotic fractures J Bone Miner Res 10: 890902.
  • 25
    Minne HW, Leidig G, Wüster C, Siromachkostov L, Baldauf G, Bickel R, Sauer P, Lojen M, Ziegler R 1988 A newly developed spine deformity index (SDI) to quantitate vertebral crush fractures in patients with osteoporosis Bone Miner 3: 335349.
  • 26
    Melton LJ, Kan SH, Frye KM, Wahner HW, O'Fallon WM, Riggs B 1989 Epidemiology of vertebral fractures in women Am J Epidemiol 129: 10001011.
  • 27
    Williams RC 1988 Toward a set of reliable and valid measures for chronic pain assessment and outcome research Pain 35: 239251.
  • 28
    Brattberg G, Thorslund M, Wikmann A 1989 The prevalence of pain in a general population: The results of a postal survey in a county of Sweden Pain 37: 215222.
  • 29
    von Korff A, Dworkin S, Le Reschel 1990 Graded chronic pain status: An epidemiologic evaluation Pain 40: 279291.
  • 30
    Zettenberg C, Mannius S, Mellstrom D, Rundgren A, Astrand K 1990 Osteoporosis and back pain in the elderly: A controlled epidemiologic and radiographic study Spine 15: 783786.
  • 31
    Bandilla K, Greif G 1987 Rückenschmerzen richtig gedeutet Diagnostica 5: 2025.
  • 32
    Sternback RA 1986 Pain hassles in the United States: Findings of the Nuprin pain report Pain 27: 6975.
  • 33
    Violinn E, Daniel L, Mc Kinney S, Loeser J 1987 When back pain becomes disabling: a regional analysis Pain 38: 3339.
  • 34
    Symmons DP, Hemert AM, Valkenburg HA 1991 A longitudinal study of back pain and radiological changes in the lumbar spines of middle-aged women Ann Rheum Dis 3: 162166.
  • 35
    Cooper C, Atkinson EJ, O'Fallon WM, Melton LJ 1992 Incidence of clinically diagnosed vertebral fractures: A population-based study in Rochester, Minnesota, 1985–1989 J Bone Miner Res 7: 221227.
  • 36
    Gold DT, Lyles KW, Bales CW, Drezuer MK 1989 Teaching patients coping behaviors: An essential part of successful management of osteoporosis J Bone Miner Res 4: 799801.
  • 37
    Agudelo CA, Semble EL, Pisko EJ, Turner RA, Anderson KO, Keefe FJ, Bradley LA, Mc Daniel LK, Young LD 1988 Prediction of pain behavior and functional status of rheumatoid arthritis patients using medical status and psychological variables Pain 33: 2532.
  • 38
    Garron DC, Leavitt F 1983 Chronic low back pain and depression J Clin Psychol 30: 486493.
  • 39
    Beutler LE, Engle D, Oro-Beutler ME, Daldrup RJ, Meredith K 1986 Inability to express intense affects: A common link between depression and pain? J Consult Clin Psychol 54: 752414.
  • 40
    Blummer D, Heilbronn M 1982 Chronic pain as a variant of depressive disease: The pain-prone disorders J Consult Clin Psychol 54: 752759.
  • 41
    Keefe FJ, Gil KM 1986 Behavioral concept in the analysis of chronic pain syndromes J Consult Clin Psychol 56: 776779.
  • 42
    Gamsa A 1990 Is emotional disturbance a precipitator or a consequence of chronic pain? Pain 42: 183195.
  • 43
    Doan BD, Wadden NP 1989 Relationships between depressive symptoms and descriptions of chronic pain Pain 36: 7584.
  • 44
    Orme JG 1986 Factorial and discriminant validity of the Center for Epidemiologic Studies of Depression J Clin Psychol 42: 2833.
  • 45
    Sinaki M, Limburg PJ, Rogers JW, Khosla S, Murtaugh PA 1990 Back muscle strength in osteoporotic compared with normal women, 1353–1354 In Christiansen C, Overgaard K (ed.) Osteoporosis 1990, Third International Symposium on Osteoporosis, Copenhagen, Denmark.
  • 46
    De Smet AA, Robinson GR, Johnson EB, Lukert BP 1988 Spinal compression fractures in osteoporotic women: Patterns and relationship to hyperkyphosis Radiology 166: 497500.
  • 47
    Itoi E, Sinaki M 1994 Effect of back-strengthening exercise on posture in healthy women 49 to 65 years of age Mayo Clin Proc 69: 10541059.