Retrieval of Individual Participant Data for Exercise Meta-Analyses May Not Be Worth the Time and Effort

Kelley, George A.; Kelley, Kristi S.

doi:https://doi.org/10.1155/2016/5059041

BioMed Research International

On this page

Abstract Introduction Methods Results Discussion Conclusions Disclosure Acknowledgments Supplementary Materials References Copyright Related Articles

Special Issue

Approaches in Physical Activity: From Basic to Applied Researches

View this Special Issue

Research Article | Open Access

Volume 2016 | Article ID 5059041 | https://doi.org/10.1155/2016/5059041

Retrieval of Individual Participant Data for Exercise Meta-Analyses May Not Be Worth the Time and Effort

George A. Kelley¹and Kristi S. Kelley¹

Academic Editor: Danilo S. Bocalini

Received05 Dec 2015

Accepted17 Feb 2016

Published08 Mar 2016

Abstract

Purpose. While individual participant data (IPD) meta-analyses are considered the gold standard for meta-analysis, the feasibility of obtaining IPD may be problematic. Methods. Using data from a previous meta-analysis of 29 studies on exercise in adults with arthritis and other rheumatic diseases, the percentage of studies in which useable IPD was provided was calculated. Results. Eight of 29 authors (28%, 95% CI = 11% to 44%) provided IPD. Using logistic regression, neither year of publication (odds ratio = 1.05, 95% CI = 0.90 to 1.27, ) nor country (odds ratio = 1.36, 95% CI = 0.20 to 10.9, ) was significantly associated with the obtainment of IPD. Conclusions. The retrieval of IPD for exercise meta-analyses may not be worth the time and effort. However, further research is needed before any final recommendations can be made.

1. Introduction

The prevalence of meta-analyses has increased dramatically over approximately the past 25 years. For example, recent PubMed searches by the first author on December 4, 2015, using the keyword “meta-analysis” found that the number of citations increased from 331 in 1990 to 14,329 in 2014 (Supplementary File 1 in Supplementary Material available online at http://dx.doi.org/10.1155/2016/5059041). The number of meta-analyses focused on exercise has also increased dramatically over the same time period. Using the keywords “exercise AND meta-analysis”, the number of meta-analyses has increased from 7 in 1990 to 472 in 2014 (Supplementary File 2). While aggregate data (AD) meta-analysis, an approach in which summary statistics from eligible studies are pooled, is still the most common type of meta-analysis, individual participant data (IPD) meta-analysis, an approach that pools the results from different studies based on data from each participant, is considered the gold standard [1–4]. However, in addition to cost [5], a major disadvantage of an IPD meta-analysis may be the ability to retrieve IPD from all eligible studies [6–8]. As a result, this negates many of the potential advantages of an IPD meta-analysis (investigating heterogeneity at the participant level, avoiding ecological bias, and harmonizing different data sources). A previous study of 142 IPD meta-analyses across a variety of different subject areas reported that the obtainment of IPD ranged from 25% to 100% [9]. With respect to exercise, the authors are only aware of one previous study regarding the obtainment of IPD. In this study, the current investigative team reported the retrieval of IPD from only 29 of 76 (38.2%) eligible studies dealing with the effects of exercise on bone mineral density in adults [10]. However, this study was conducted approximately 13 years ago. Since that time, technological advances have improved one’s ability to store and share data, thereby making it easier to share deidentified IPD with others. Given that meta-analyses are increasingly used to provide evidence-based decisions regarding the benefits of exercise on a variety of health outcomes, a basic understanding of the best approach for conducting such analyses is needed. Therefore, the purpose of this brief report was to contribute to the scant exercise literature by providing up-to-date information regarding the retrieval of IPD, including potential covariates, for a specific meta-analysis.

2. Methods

2.1. Data Source

Data were derived from a recently published AD exercise meta-analysis in which an IPD meta-analysis was originally planned, details of which have been described elsewhere [11]. Briefly, the specific aims were to calculate a summary effect of exercise on depressive symptoms in adults with arthritis and other rheumatic diseases and explore for potential sources of heterogeneity. Studies were included if they were randomized controlled trials examining the effects of exercise (aerobic, strength training, or both) on depressive symptoms in adults with arthritis and other rheumatic diseases [11]. Twenty-nine studies representing 2,449 participants (1,470 exercise, 979 control) were included [12–40].

2.2. Retrieval of IPD

Using a predefined form letter (see Supplementary File 3), deidentified IPD was requested by having the second author contact the corresponding author of each eligible study via electronic mail asking if they would be interested in providing IPD. A response was requested within two weeks, regardless of interest, with the authors being informed that they would be listed in the acknowledgements section of each published study derived from the project if they provided their IPD. If no response was received within two weeks, up to two additional requests were sent via electronic mail. For those who responded but chose not to contribute, reasons given for not participating, if any, were recorded. For those authors who expressed interest in providing IPD, a second electronic mail was sent that included an attachment consisting of a predefined list of IPD requested (see Supplementary File 4). Investigators were asked to provide IPD in a format that was convenient for them within four weeks. If IPD was not received within four weeks, as many as two additional reminders were sent via electronic mail. The dates of all communications were recorded.

2.3. Statistical Analysis

Descriptive statistics were used to describe the number of responses to initial electronic mail requests for IPD, number of days to respond to initial electronic mail requests, number of authors who provided useable IPD, and number of days from initial requests to receipt of IPD. Reasons for not providing IPD from authors who were willing to supply such information were also recorded. Furthermore, descriptive statistics were calculated for the two potential predictors included in the regression model.

Because of the small sample size, exact logistic regression was used to examine potential predictors with respect to whether or not IPD was received [41]. Based on the investigative team’s previous research [10], the two potential predictors included in the model were country in which the study was conducted and year that the study was published. The chi-square distribution () was used to examine the overall model. The alpha level for statistical significance was set at .

3. Results

3.1. Descriptive Statistics for Retrieval of IPD

The authors from 20 of 29 studies (69.0%, 95% CI = 52.2% to 85.8%) responded to initial electronic mail requests for IPD while 9 (31.0%, 95% CI = 14.2% to 47.8%) never responded despite multiple requests. The response time to initial requests varied widely from 1 to 181 days ( = 54.2 ± 74.8, 95% CI = 21.4 to 87.0, Mdn = 17). Eight of 29 authors (27.6%, 95% CI = 11.3% to 43.8%) provided IPD. The number of days from initial requests for data to receipt of IPD ranged from 36 to 179 ( = 74.4 ± 46.4, 95% CI = 42.0 to 106.8, Mdn = 64). Reasons for not providing IPD included no longer having the data () and time (). Another author said that they would supply IPD if a consortium was formed and in which they were included as a coauthor. Year of publication ranged from 1989 to 2011. Thirteen studies (44.8%, 95% CI = 26.7% to 62.9%) were conducted in the United States while the remaining 16 (55.2%, 95% CI = 37.1% to 73.3%) were conducted in countries other than the United States.

3.2. Potential Predictors in the Obtainment of IPD

Results for exact logistic regression are shown in Table 1. The overall model was not statistically significant ( 0.62, ) and neither year of publication nor country was a significant predictor for the receipt of IPD ( for both).

4. Discussion

4.1. Overall Findings

The current study resulted in a low obtainment of IPD for this exercise meta-analysis, with less than one-third of eligible studies providing IPD. The response rates observed are either similar to [10, 42], lower than, [9, 43] or higher [44] than previous research across a variety of different subject areas. While one might speculate that the formation of a consortium could have increased the response rate, a previous IPD meta-analysis on obesity and diabetes was only able to retrieve approximately 40% of IPD when such a consortium was established [42]. The lack of association between year and country with the obtainment of IPD is in contrast with previous research where a trend was found for both to be associated with the retrieval of IPD [10]. One possible reason for this discrepancy may have been the lack of statistical power because of the smaller number of studies included (29 versus 76) [10]. Finally, and as previously reported in the original meta-analysis [11], no statistically significant or clinically important differences were found in depressive symptoms between those studies that supplied IPD (standardized mean difference (SMD), −0.31, 95% CI, −0.61 to −0.01) and those that did not (SMD, −0.47, 95% CI, −0.65 to −0.28), a finding consistent with the authors’ previous research [10].

4.2. Implications for Research and Practice

The study reinforces the importance of considering the feasibility of obtaining IPD when deciding whether to conduct an AD or IPD exercise meta-analysis. In addition, the increased costs associated with an IPD meta-analysis need to be considered. For example, Steinberg et al. estimated that the costs associated with conducting a meta-analysis of 12 ovarian cancer studies were more than 5 times greater using the IPD versus AD approach [5]. However, others estimated the costs of this same study to be at least 8 times greater given that the investigative team continued to work on the study after funding for the project ended [45].

4.3. Strengths and Potential Limitations

This study provides important up-to-date information regarding the retrieval of IPD for an exercise meta-analysis. This is noteworthy given that this is only the second study that has examined the feasibility of retrieving IPD for an exercise meta-analysis, the first being conducted by the authors back in 2002 [10]. In addition, given that meta-analyses are increasingly used to provide evidence-based decisions regarding the benefits of exercise on a plethora of health outcomes, the results of this study provide critically important information regarding the feasibility of conducting an IPD exercise meta-analysis. However, since the current study focused on one attempted IPD meta-analysis, the findings may not be generalizable to other IPD exercise meta-analyses. In addition, no cost data were collected or analyzed, thereby limiting the applicability of results. Furthermore, the collapsing of countries other than the US into one category because of the small number of studies available for each country could have biased the results. Finally, a better approach for examining potential differences between AD and IPD may have been to include a mixture of AD and IPD rather than dichotomizing the two and comparing them using an AD approach [9, 46]. However, the investigative team was not comfortable using these methods given the inability to obtain the data necessary to replicate the statistical methods and results reported in this previous work (Riley, written communication, July 11, 2012; Staessen, written communication, July 30, 2012) [46].

5. Conclusions

The retrieval of IPD for exercise meta-analyses may not be worth the time and effort. However, further research is needed before any final recommendations can be made.

Disclosure

Kristi S. Kelley is a coauthor of this paper. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Competing Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

Acknowledgments

This work was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health (Grant no. RO1 AR061346) and the National Institute of General Medical Sciences of the National Institutes of Health (Grant no. U54 GM104942). The authors would like to thank the following people for supplying their IPD: Susan Buckelew, Jennifer L. Etnier, Marlene Fransen, A. Hakkinen, Geri B. Neuberger, Sheila C. O’Reilly, Michael Doherty, Joanna Romawski, Mike E. Miller, Brenda Penninx, William B. Applegate, Candace L. Schachter, Ayse Karan, Sureyya Sencan, Jamil Natour, Valerie Valim, and H. Valkeinen.

Supplementary Materials

Supplementary File 1: This file provides details regarding a PubMed search conducted on December 4, 2015 for citations indexed between the years 1990 and 2014 using the key term “meta-analysis”.

Supplementary File 2: This file provides details regarding a PubMed search conducted on December 4, 2015 for citations indexed between the years 1990 and 2014 using the key terms “exercise and meta-analysis”.

Supplementary File 3: This file is an example of the initial letter that was sent to authors requesting de-identified individual participant data for their previously published study.

Supplementary File 4: This file is an example of the follow-up letter that was sent to eligible investigators describing what de-identified individual participant data was needed from their previously published study.

Supplementary Material

References

R. D. Riley, P. C. Lambert, and G. Abo-Zaid, “Meta-analysis of individual participant data: rationale, conduct, and reporting,” British Medical Journal, vol. 340, article c221, 2010.
View at: Publisher Site | Google Scholar
R. D. Riley, “Commentary: Like it and lump it? Meta-analysis using individual participant data,” International Journal of Epidemiology, vol. 39, no. 5, pp. 1359–1361, 2010.
View at: Publisher Site | Google Scholar
R. D. Riley, I. Kauser, M. Bland et al., “Meta-analysis of randomised trials with a continuous outcome according to baseline imbalance and availability of individual participant data,” Statistics in Medicine, vol. 32, no. 16, pp. 2747–2766, 2013.
View at: Publisher Site | Google Scholar | MathSciNet
L. A. Stewart and M. K. B. Parmar, “Meta-analysis of the literature or of individual patient data: is there a difference?” The Lancet, vol. 341, no. 8842, pp. 418–422, 1993.
View at: Publisher Site | Google Scholar
K. K. Steinberg, S. J. Smith, D. F. Stroup et al., “Comparison of effect estimates from a meta-analysis of summary data from published studies and from a meta-analysis using individual patient data for ovarian cancer studies,” American Journal of Epidemiology, vol. 145, no. 10, pp. 917–925, 1997.
View at: Publisher Site | Google Scholar
T. P. Debray, K. G. Moons, G. van Valkenhoef et al., “Get real in individual participant data (IPD) meta-analysis: a review of the methodology,” Research Synthesis Methods, vol. 6, no. 4, pp. 293–309, 2015.
View at: Publisher Site | Google Scholar
J. F. Tierney, C. Vale, R. Riley et al., “Individual participant data (IPD) meta-analyses of randomised controlled trials: guidance on their use,” PLoS Medicine, vol. 12, no. 7, Article ID e1001855, 2015.
View at: Publisher Site | Google Scholar
P. Saramago, L.-H. Chuang, and M. O. Soares, “Network meta-analysis of (individual patient) time to event data alongside (aggregate) count data,” BMC Medical Research Methodology, vol. 14, article 105, 2014.
View at: Publisher Site | Google Scholar
R. D. Riley, M. C. Simmonds, and M. P. Look, “Evidence synthesis combining individual patient data and aggregate data: a systematic review identified current practice and possible methods,” Journal of Clinical Epidemiology, vol. 60, no. 5, pp. 431.e1–439.e12, 2007.
View at: Publisher Site | Google Scholar
G. A. Kelley, K. S. Kelley, and Z. V. Tran, “Retrieval of individual patient data for an exercise meta-analysis,” American Journal of Medicine and Sports, vol. 4, pp. 350–354, 2002.
View at: Google Scholar
G. A. Kelley, K. S. Kelley, and J. M. Hootman, “Effects of exercise on depression in adults with arthritis: a systematic review with meta-analysis of randomized controlled trials,” Arthritis Research & Therapy, vol. 17, article 21, 2015.
View at: Publisher Site | Google Scholar
E. Alentorn-Geli, J. Padilla, G. Moras, C. L. Haro, and J. Fernández-Solà, “Six weeks of whole-body vibration exercise improves pain and fatigue in women with fibromyalgia,” Journal of Alternative and Complementary Medicine, vol. 14, no. 8, pp. 975–981, 2008.
View at: Publisher Site | Google Scholar
S. P. Buckelew, R. Conway, J. Parker et al., “Biofeedback/relaxation training and exercise interventions for fibromyalgia: a prospective trial,” Arthritis & Rheumatism, vol. 11, no. 3, pp. 196–209, 1998.
View at: Publisher Site | Google Scholar
L. H. Daltroy, C. Robb-Nicholson, M. D. Iversen, E. A. Wright, and M. H. Liang, “Effectiveness of minimally supervised home aerobic training in patients with systemic rheumatic disease,” British Journal of Rheumatology, vol. 34, no. 11, pp. 1064–1069, 1995.
View at: Publisher Site | Google Scholar
J. L. Etnier, W. B. Karper, J. I. Gapin, L. A. Barella, K. C. Yu, and K. J. Murphy, “Exercise, fibromyalgia, and fibrofog: a pilot study,” Journal of Physical Activity & Health, vol. 6, no. 2, pp. 239–246, 2009.
View at: Google Scholar
M. Fransen, L. Nairn, J. Winstanley, P. Lam, and J. Edmonds, “Physical activity for osteoarthritis management: a randomized controlled clinical trial evaluating hydrotherapy or Tai Chi classes,” Arthritis Care & Research, vol. 57, no. 3, pp. 407–414, 2007.
View at: Publisher Site | Google Scholar
K. R. Fontaine, L. Conn, and D. J. Clauw, “Effects of lifestyle physical activity on perceived symptoms and physical function in adults with fibromyalgia: results of a randomized trial,” Arthritis Research and Therapy, vol. 12, article R55, 2010.
View at: Publisher Site | Google Scholar
S. E. Gowans, A. dehueck, S. Voss, A. Silaj, S. E. Abbey, and W. J. Reynolds, “Effect of a randomized, controlled trial of exercise on mood and physical function in individuals with fibromyalgia,” Arthritis Care and Research, vol. 45, no. 6, pp. 519–529, 2001.
View at: Publisher Site | Google Scholar
T. Haak and B. Scott, “The effect of Qigong on Fibromyalgia (FMS): a controlled randomized study,” Disability and Rehabilitation, vol. 30, no. 8, pp. 625–633, 2008.
View at: Publisher Site | Google Scholar
A. Häkkinen, K. Häkkinen, P. Hannonen, and M. Alen, “Strength training induced adaptations in neuromuscular function of premenopausal women with fibromyalgia: comparison with healthy women,” Annals of the Rheumatic Diseases, vol. 60, no. 1, pp. 21–26, 2001.
View at: Publisher Site | Google Scholar
M. R. Ide, L. M. M. Laurindo, A. L. Rodrigues-Júnior, and C. Tanaka, “Effect of aquatic respiratory exercise-based program in patients with fibromyalgia,” International Journal of Rheumatic Diseases, vol. 11, no. 2, pp. 131–140, 2008.
View at: Publisher Site | Google Scholar
K. D. Jones, C. S. Burckhardt, A. A. Deodhar, N. A. Perrin, G. C. Hanson, and R. M. Bennett, “A six-month randomized controlled trial of exercise and pyridostigmine in the treatment of fibromyalgia,” Arthritis and Rheumatism, vol. 58, no. 2, pp. 612–622, 2008.
View at: Publisher Site | Google Scholar
G. R. Komatireddy, R. W. Leitch, K. Cella, G. Browning, and M. Minor, “Efficacy of low load resistive muscle training in patients with rheumatoid arthritis functional class II and III,” Journal of Rheumatology, vol. 24, no. 8, pp. 1531–1539, 1997.
View at: Google Scholar
K. Mannerkorpi, B. Nyberg, M. Ahlmen, and C. Ekdahl, “Pool exercise combined with an education program for patients with fibromyalgia syndrome. A prospective, randomized study,” Journal of Rheumatology, vol. 27, no. 10, pp. 2473–2481, 2000.
View at: Google Scholar
M. A. Minor, J. E. Hewett, R. R. Webel, S. K. Anderson, and D. R. Kay, “Efficacy of physical conditioning exercise in patients with rheumatoid arthritis and osteoarthritis,” Arthritis and Rheumatism, vol. 32, no. 11, pp. 1396–1405, 1989.
View at: Publisher Site | Google Scholar
G. B. Neuberger, L. S. Aaronson, B. Gajewski et al., “Predictors of exercise and effects of exercise on symptoms, function, aerobic fitness, and disease outcomes of rheumatoid arthritis,” Arthritis Care and Research, vol. 57, no. 6, pp. 943–952, 2007.
View at: Publisher Site | Google Scholar
S. C. O'Reilly, K. R. Muir, and M. Doherty, “Effectiveness of home exercise on pain and disability from osteoarthritis of the knee: a randomised controlled trial,” Annals of the Rheumatic Diseases, vol. 58, no. 1, pp. 15–19, 1999.
View at: Publisher Site | Google Scholar
D. L. Patrick, S. D. Ramsey, A. C. Spencer, S. Kinne, B. Belta, and T. D. Topolski, “Economic evaluation of aquatic exercise for persons with osteoarthritis,” Medical Care, vol. 39, no. 5, pp. 413–424, 2001.
View at: Publisher Site | Google Scholar
B. W. J. H. Penninx, W. J. Rejeski, J. Pandya et al., “Exercise and depressive symptoms: a comparison of aerobic and resistance exercise effects on emotional and physical function in older persons with high and low depressive symptomatology,” Journals of Gerontology B: Psychological Sciences and Social Sciences, vol. 57, no. 2, pp. P124–P132, 2002.
View at: Publisher Site | Google Scholar
D. S. Rooks, S. Gautam, M. Romeling et al., “Group exercise, education, and combination self-management in women with fibromyalgia: a randomized trial,” Archives of Internal Medicine, vol. 167, no. 20, pp. 2192–2200, 2007.
View at: Publisher Site | Google Scholar
B. Sañudo, D. Galiano, L. Carrasco, M. de Hoyo, and J. G. McVeigh, “Effects of a prolonged exercise program on key health outcomes in women with fibromyalgia: a randomized controlled trial,” Journal of Rehabilitation Medicine, vol. 43, no. 6, pp. 521–526, 2011.
View at: Publisher Site | Google Scholar
C. L. Schachter, A. J. Busch, P. M. Peloso, and M. S. Sheppard, “Effects of short versus long bouts of aerobic exercise in sedentary women with fibromyalgia: a randomized controlled trial,” Physical Therapy, vol. 83, no. 4, pp. 340–358, 2003.
View at: Google Scholar
S. Sencan, S. Ak, A. Karan, L. Muslumanoglu, E. Ozcan, and E. Berker, “A study to compare the therapeutic efficacy of aerobic exercise and paroxetine in fibromyalgia syndrome,” Journal of Back and Musculoskeletal Rehabilitation, vol. 17, no. 2, pp. 57–61, 2004.
View at: Google Scholar
P. Tomas-Carus, N. Gusi, A. Leal, Y. García, and A. Ortega-Alonso, “The fibromyalgia treatment with physical exercise in warm water reduces the impact of the disease on female patients' physical and mental health,” Reumatología Clínica, vol. 3, no. 1, pp. 33–37, 2007.
View at: Publisher Site | Google Scholar
P. Tomas-Carus, N. Gusi, A. Häkkinen, K. Häkkinen, A. Leal, and A. Ortega-Alonso, “Eight months of physical training in warm water improves physical and mental health in women with fibromyalgia: a randomized controlled trial,” Journal of Rehabilitation Medicine, vol. 40, no. 4, pp. 248–252, 2008.
View at: Publisher Site | Google Scholar
V. Valim, L. Oliveira, A. Suda et al., “Aerobic fitness effects in fibromyalgia,” Journal of Rheumatology, vol. 30, no. 5, pp. 1060–1069, 2003.
View at: Google Scholar
H. Valkeinen, M. Alen, P. Hannonen, A. Häkkinen, O. Airaksinen, and K. Häkkinen, “Changes in knee extension and flexion force, EMG and functional capacity during strength training in older females with fibromyalgia and healthy controls,” Rheumatology, vol. 43, no. 2, pp. 225–228, 2004.
View at: Publisher Site | Google Scholar
C. Wang, C. H. Schmid, P. L. Hibberd et al., “Tai Chi is effective in treating knee osteoarthritis: a randomized controlled trial,” Arthritis Care and Research, vol. 61, no. 11, pp. 1545–1553, 2009.
View at: Publisher Site | Google Scholar
C. Wang, C. H. Schmid, R. Rones et al., “A randomized trial of tai chi for fibromyalgia,” The New England Journal of Medicine, vol. 363, no. 8, pp. 743–754, 2010.
View at: Publisher Site | Google Scholar
S. H. Wigers, T. C. Stiles, and P. A. Vogel, “Effects of aerobic exercise versus stress management treatment in fibromyalgia. A 4.5 year prospective study,” Scandinavian Journal of Rheumatology, vol. 25, no. 2, pp. 77–86, 1996.
View at: Publisher Site | Google Scholar
C. R. Mehta and N. R. Patel, “Exact logistic regression: theory and examples,” Statistics in Medicine, vol. 14, no. 19, pp. 2143–2160, 1995.
View at: Publisher Site | Google Scholar
S. Duval, G. Vazquez, W. L. Baker, and D. R. Jacobs Jr., “The Collaborative Study of Obesity and Diabetes in Adults (CODA) project: meta-analysis design and description of participating studies,” Obesity Reviews, vol. 8, no. 3, pp. 263–276, 2007.
View at: Publisher Site | Google Scholar
M. F. Piepoli, C. Davos, D. P. Francis, and A. J. Coats, “Exercise training meta-analysis of trials in patients with chronic heart failure (ExTraMATCH),” British Medical Journal, vol. 328, article 189, 2004.
View at: Publisher Site | Google Scholar
B. Villain, A. Dechartres, P. Boyer, and P. Ravaud, “Feasibility of individual patient data meta-analyses in orthopaedic surgery,” BMC Medicine, vol. 13, article 131, 2015.
View at: Publisher Site | Google Scholar
H. Cooper and E. A. Patall, “The relative benefits of meta-analysis conducted with individual participant data versus aggregated data,” Psychological Methods, vol. 14, no. 2, pp. 165–176, 2009.
View at: Publisher Site | Google Scholar
R. D. Riley, P. C. Lambert, J. A. Staessen et al., “Meta-analysis of continuous outcomes combining individual patient data and aggregate data,” Statistics in Medicine, vol. 27, no. 11, pp. 1870–1893, 2008.
View at: Publisher Site | Google Scholar | MathSciNet

Copyright

Copyright © 2016 George A. Kelley and Kristi S. Kelley. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

1369

Downloads

773

Citations