Adjusting for Recall Bias with the Proportional Hazards Model

In studying falling frequency in the elderly, we observed that having subjects keep a diary led to a larger number of falls reported than had been noted in a previous study in the same population. The previous study asked subjects to report any falls in the previous three months. We considered two related explanations for the observation of lower incidence reports with a 3-month recall survey. First, there may have been under-reporting of falls due to recall bias. Second, the less severe falls (which did not result in injuries) may not be reported. We suggest that the proportional hazards model may be used to adjust studies in which recall is used to determine incidence and time to falls.

Interexaminer reliability of the palpation of trigger points in the trunk and lower limb muscles

OBJECTIVES

To determine the interexaminer reliability of palpation of three characteristics of trigger points (taut band, local twitch response, and referred pain) in patients with subacute low back pain, to determine whether training in palpation would improve reliability, and whether there was a difference between the physiatric and chiropractic physicians.

DESIGN

Reliability study.

SETTING

Whittier Health Campus, Los Angeles College of Chiropractic.

PARTICIPANTS

Twenty-six nonsymptomatic individuals and 26 individuals with subacute low back pain.

INTERVENTION

Twenty muscles per individual were first palpated by an expert and then randomly by four physician examiners.

MAIN OUTCOME MEASURES

Palpation findings.

RESULTS

Kappa scores for palpation of taut bands, local twitch responses, and referred pain were .215, .123, and .342, respectively, between the expert and the trained examiners, and .050, .118, and .326, respectively, between the expert and the untrained examiners. Kappa scores for agreement for palpation of taut bands, twitch responses, and referred pain were .108, -.001, and .435, respectively, among the nonexpert, trained examiners, and -.019, .022, and .320, respectively, among the nonexpert, untrained examiners.

CONCLUSIONS

Among nonexpert physicians, physiatric or chiropractic, trigger point palpation is not reliable for detecting taut band and local twitch response, and only marginally reliable for referred pain after training.

Attempts to prevent falls and injury: a prospective community study

At 16 senior centers, we studied the effectiveness of exercise and cognitive-behavioral programs, compared with a discussion control program, in reducing falls and injuries among 230 older adults. After 1 year of the programs, we observed no significant difference in time to first fall. Even though a relatively high percentage (38.6%) suffered at least one fall, only 7.8% of these community-residing elderly required medical attention. Secondary outcome measures such as strength, balance, fear of falling, and perceived health did not significantly change.

Intergenerational relations: Pre-med students at senior centers

Misconceptions about aging and a negative bias against caring for older adults entail unfortunate consequences for society. This study explored the potential of frequent personal contact in reducing misconceptions and improving personal attitude towards older adults. Thirty students, predominantly Asian, with English as their second language, attended senior centers three times per week for 11 months, on the average. They assisted in teaching health promotion classes for the prevention of falls, conducted health assessments by interview, and collected physiological data. At the senior centers, a sample of 70 older adult Americans participated in the health promotion classes and in this evaluation of intergenerational relations. Students completed anonymously two questionnaires; one was objective, testing misconceptions about aging; the other was subjective, rating their reactions to elders at senior centers. Older adults completed, also anonymously, a subjective questionnaire on their reactions to students at their centers. Results on students' misconceptions about aging were similar to those among students at Duke University who did not experience frequent personal contact with well elderly. According to the Facts on Aging Quiz, the overall attitude among students at senior centers was more biased in the negative direction than among Duke students, especially among students who had just joined the project. However, in the subjective questionnaires, students at senior centers expressed feeling at ease at the centers and they disagreed with statements of impatience and frustration. Older adults at the centers had a positive attitude toward the students and most thought that students were sympathetic.

Adjusting for recall bias with the proportional hazards model

In studying falling frequency in the elderly, we observed that having subjects keep a diary led to a larger number of falls reported than had been noted in a previous study in the same population. The previous study asked subjects to report any falls in the previous three months. We considered two related explanations for the observation of lower incidence reports with a 3-month recall survey. First, there may have been under-reporting of falls due to recall bias. Second, the less severe falls (which did not result in injuries) may not be reported. We suggest that the proportional hazards model may be used to adjust studies in which recall is used to determine incidence and time to falls.

Falling among the sensorially impaired elderly

Recent studies of the healthy elderly indicate that accidental falls account for many lengthy hospital stays and permanent disabilities. An ongoing investigation by our group is examining the contribution of neuromuscular feedback and the possible substitution of visual-perceptual feedback as a major contributor to falling in this population. This combination of decreased weighting of neuromuscular factors and increased weighting of the visual-perceptual factors led to the following hypothesis: Those persons who cannot see should show less change in frequency of falling as they age than sighted cohorts. The deaf, who rely more than others on visual perceptual stimuli, are less likely to fall as they age than the general elderly population, who first begin to rely more heavily on visual stimuli when they age. Thus, the deaf and blind constitute a population whose change in frequency of falls with increased age should differ markedly from change in the frequency of falls among the nonblind and nondeaf segments of the population. Our results support the hypothesis. The blind demonstrated a higher rate of falls than the deaf or nonimpaired population. However, the more elderly blind failed to show the increase in falling demonstrated by the deaf and nonimpaired. If these findings are confirmed in subsequent studies, the nonimpaired and deaf elderly could be trained to focus on visual feedback. The blind may be able to reduce the frequency of falls by enhancing musculoskeletal feedback and strength via exercise.

Visual perception dominance of fallers among community-dwelling older adults

The authors postulated that older adult fallers show a greater tendency than older adult nonfallers to rely more on visual information sources in maintaining upright posture than on kinesthetic and vestibular cues. This paper presents descriptive statistics on 199 older adults living independently in the community. Their visual perception of the vertical and horizontal was analyzed with respect to age, sex, health status, and severity of injury as a result of a fall. The finding of significant impairments for fallers in visual perceptual abilities confirmed a trend previously established by one of the authors (Tobis). When the visual field entailed only misleading or ambiguous cues in the form of a tilted frame, fallers again showed a larger error than nonfallers in establishing the vertical and horizontal. The authors feel that this relatively greater dependence on visual sources may develop in response to impairment of feedback on posture and gait from the kinesthetic and vestibular systems as a result of age and chronic health problems. Errors in visual perception of the vertical and horizontal intercorrelated with age, sex, and a large number of medical problems. However, visual variables were more important in predicting faller status than physical characteristics.

Psychological factors in the treatment of back pain by spinal manipulation

Ninety patients with low back pain completed a short form of the Minnesota Multiphasic Personality Inventory (MMPI) and, subsequently, received spinal manipulative therapy. Most (84%) of the patients reported immediate relief following manipulation but the extent of that relief was not correlated with any psychological measure. However, several days after treatment, the percentage of patients reporting improvement was much lower (55%) and lack of improvement was associated with elevated scores on hypochondriasis (Hs), and functional low back pain (Lb). Unimproved patients also tended to report longer durations of back pain prior to treatment. Multiple regression analyses indicated that pain duration and the psychological factors were independent predictors of the response to treatment. It is suggested that, although a single manipulative treatment will often relieve the immediate cause of back pain, underlying psychosomatic factors may predispose the condition to recur.

Low back pain and its treatment by spinal manipulation: measures of flexibility and asymmetry

Nineteen low back pain patients and eight patients not suffering from low back pain were given several tests of flexibility and asymmetry by two different examiners. Three criteria of reliability and validity were used: (1) significant agreement between independent observers, (2) significantly different scores in the groups with and without low back pain, and (3) significant improvement following a successful spinal manipulation. Tests of anterior flexion and asymmetry of foot eversion met only the first and second criteria while tests of hamstring tightness and asymmetry of voluntary straight leg raising met only the first and third criteria. Passive and voluntary straight leg raising tests were the only measures that met all three criteria. Therefore, of the objective tests investigated here, only passive or voluntary straight leg raising can be strongly recommended for use in the evaluation of spinal manipulative therapy for low back pain.

Visual perception of verticality and horizontality among elderly fallers

Patients who had suffered repeated falls or a single fall which led to hospitalization were tested for errors in visual perception of the verticality or horizontality of a rod presented on a video screen. Fallers had significantly greater average error scores than nonfallers. Fallers were more likely to be elderly or hemiplegic, and these conditions were also associated with elevated error scores. However, when age or hemiplegia was removed as a factor in the analysis, perceptual errors were still significantly greater in fallers than in nonfallers.

Spinal manipulation for low back pain

A randomized clinical trial of rotational manipulation was conducted on 95 patients with low back pain selected for (1) the absence of any contraindications for vertebral manipulation, (2) the absence of any psychosocial problems that might affect the outcome of treatment, (3) the absence of any previous experience with manipulative therapy, and (4) the presence of palpatory cues indicating that manipulation might be successful. Patients were randomly assigned to one of two groups: an experimental group receiving manipulation therapy and a control group receiving soft-tissue massage. Comparison of the two groups indicated that (1) patients who received manipulative treatment were much more likely to report immediate relief after the first treatment, and (2) at discharge, there was no significant difference between the two groups because both showed substantial improvement.

Rehabilitation recommendations for institutionalized motor-disabled patients

Consultation reports on 142 profoundly retarded and severely motor-disabled patients in a state residential care facility who were seen by physiatrists from an academic department of physical medicine and rehabilitation (PM&R) were reviewed. In all, 168 reasons for consultation were listed: 117 for specific identification or management of disabilities related to gross-motor or fine-motor function, 37 for surgical problems, and 10 for medical problems. The consultants made 314 recommendations altogether: 100 suggested the evaluation or treatment of gross-motor disabilities by a physical therapist, 43 suggested occupational therapy evaluation or treatment, and an additional 73 suggested adaptive equipment involving an orthotist. The recommendations also involved the following problems: medical-33; surgical-37; behavioral-27; further rehabilitation-1. The skills of a consultant knowledgeable in the management of multiply handicapped people include the coordination of medical, surgical, physical and behavioral care. Such knowledge is also a useful adjunct to overall patient care in complex cases with motor involvement. The preliminary association during which these consultations were done served as the basis of a growing affiliation between the state institution and the department of PM&R.