• The Effect of Three-year Weight History on Quantitative Ultrasound Parameters in Older Mainland Chinese Women and Men

      Sanchez, Ana Maria; Magaziner, Jay (2015)
      Title: The Effect of Three-year Weight History on Quantitative Ultrasound Parameters in Older Mainland Chinese Women and Men Background: Finding risk factors for poor bone quality that are relevant to China is useful for targeting higher-risk women for bone quantitative ultrasound (QUS) testing. Weight history has typically been measured as the difference in weight between two time points. Incorporating an additional indicator of weight variability over this time period may enhance this measure. Objective: To examine the relationship between bone QUS and weight history (including weight variability) in pre and postmenopausal women, women transitioning into menopause (TIM) and men over 65 years of age. Methods: This was a retrospective, clinic-based study of 1201 postmenopausal, 862 pre-menopausal and 419 TIM women, and 424 older men in Guangdong Province, China. Weight was measured yearly over a three year period. Weight history over 3 years was defined as (1) weight trend (the slope of weight on year) and weight variability (the root mean square error (RMSE) around the slope of weight on year) (2) weight trajectories (3-year trajectories estimated by growth mixture modeling), and (3) weight change (the percent change in weight from baseline). Stiffness index (SI) of the os calcaneus [assessed using a Lunar Achilles Insight QUS device (GE Medical Systems Inc., USA)], age, height and weight were measured once at the end of year 3. Z-scores were used to compare regression lines for predicting SI from the weight history variables among study groups. Results: Weight variability (weight-RMSE) was significantly and negatively associated with SI, independent of weight, weight trend, age and height (β = -1.32; P< 0.03) among postmenopausal women. No other associations between weight history and SI were detected. Slopes for the association between SI and weight-RMSE were not significantly different between postmenopausal women and the other study groups. Conclusions: This study provides evidence that measuring weight variability, in addition to weight change, may provide a more complete assessment of the influence of weight history on bone. Further work is needed to investigate methods of quantifying weight variability that could be used in a clinical setting.
    • Evaluating the Relationship between Muscle and Bone Modeling Response in Older Adults

      Reider, Lisa; Magaziner, Jay (2014)
      Background: Bone modeling, the process that continually adjusts bone strength in response to prevalent muscle-loading forces throughout an individual's lifespan, may play an important role in bone fragility with age. Femoral stress, an index of bone modeling response can be estimated using measurements of DXA derived bone geometry and loading information incorporated into an engineering model. Assuming that individuals have adapted to habitual muscle loading forces, greater stresses indicate a diminished response and a weaker bone Aims/Methods: The aims of this dissertation were to 1) evaluate the association of femoral stress with measures of lean mass and muscle strength among healthy older adults participating in the Health ABC study using linear regression; 2) determine whether femoral stress predicts incident fracture among the same cohort of older adults using cox proportional hazards models; and 3) evaluate the association of femoral stress with measures of lean mass and muscle strength in women after hip fracture participating in the 3rd and 4th cohort of the Baltimore Hip Studies using linear regression and to determine whether femoral stress changes the year following fracture using longitudinal data analysis. Results: Lean mass explained more of the variation in femoral stress than measures of muscle strength among healthy older men and women as well as in women with hip fracture. Remaining variability in femoral stress may reflect individual variation in modeling response. After adjusting for measures of lean mass and strength, women in the highest tertile of femoral stress had 77% higher hazard of fracture and men in the highest tertile of femoral stress had 84% higher hazard of fracture relative to women and men in the lowest tertile, respectively. This suggests that deficiencies in bone modeling response may be an important predictor of fracture. Femoral stress did not appear to change the year following fracture in older women. Conclusion: Future studies should focus on refining measures of bone modeling response by incorporating better measures of muscle force. While femoral stress does not have clinical applications per se, it allows us to investigate a potentially important mechanism underlying bone fragility and provides a framework for thinking about treatments that could improve the interaction between muscle and bone.
    • Inappropriate medication use and transition to nursing home among community-dwelling elders

      Zuckerman, Ilene H.; Magaziner, Jay (2005)
      Background. The objective of this observational study was to investigate the relationship between transitions from community to nursing home (NH) and inappropriate medication use in older adults. Specific aims were to estimate: (1) incidence of transitions from the community to NH; (2) prevalence of inappropriate medication use among community-dwelling elders; and (3) the association between transitions from the community to NH and inappropriate medication use. Methods. A retrospective cohort was assembled from a sample of elderly Medicare beneficiaries with employer-sponsored supplemental health insurance. A longitudinal data set was created from Medicare supplemental insurance claims at the person-month level. After an initial baseline year with no NH events, each subject was followed until the first month of NH transition, loss to follow-up or the end of the 24-month follow-up period. The exposure was a time-dependent variable defined as use of any one of twenty-two categories of drugs considered by experts to be inappropriate in the elderly. Survival analysis for interval-censored data was used to compare the hazard of NH transition among those with and without inappropriate drug use in the previous three months. Results. 487,383 subjects were eligible for inclusion. Mean age was 74 years; 56% were female. Annual prevalence of any inappropriate drug was 42%. NH incidence rate was 28.4 per 1,000 person-years. After adjustment for potential confounders, the use of any inappropriate drug was associated with a 31% increase in risk of transition, compared to no use of inappropriate drug, within the past three months (adjusted hazard ratio 1.31, 99% CI 1.26, 1.36). Analyses of individual drug classes showed the risk of nursing home transition was similar for inappropriate drugs and alternative drugs of the same class. Conclusion. Inappropriate drug use was associated with an increased risk of nursing home transition. However, findings suggest that the association may be confounded by indication. That is, it may not be the "inappropriateness" of the drug, but rather the indication for the drug itself, that increased the risk of NH transition. However, use of these medications may be a marker that could be used to identify people at risk for NH transition.
    • PICTURE (Psychoeducational Intervention for Children of Trauma and Understanding Through Research and Education)

      Reynolds, Matthew William; Magaziner, Jay (2000)
      The Psychoeducational Intervention for Children of Trauma by Understanding through Research and Education (PICTURE) evaluated whether an art psychotherapy intervention was effective in improving short-term psychological and behavioral outcomes in elementary school children who have witnessed domestic violence. The study population consisted of 45 children, aged 5-12, from three elementary schools in Baltimore County, Maryland, who had witnessed domestic violence. Children were randomized to either an art therapy intervention followed by an arts and crafts class or an arts and crafts class followed by art therapy. Each intervention lasted for 10 sessions and met once a week for 45 to 60 minutes. Information was collected from the children on self-esteem and depression, and from their teachers on the children's behaviors. The information was collected at three time points (baseline, 10 weeks, and 20 weeks). Mixed model analyses showed that the art therapy intervention was significantly more effective than the arts and crafts class in reducing Attention Deficit-Hyperactivity Disorder risk and hyperactivity as measured by the Conners Teacher Rating Scale. There was also suggestive evidence of art therapy effectiveness in reducing depressive symptoms, teacher-rated hyperactivity and teacher-rated oppositional behavior in these elementary aged-children who had witnessed domestic violence. The study results suggest that art therapy may be a viable option toward helping children who have witnessed domestic violence with the resulting psychological symptoms and behaviors.