Limits...
An assessment of the characteristics and quality of diagnostic accuracy studies for positron emission tomography conducted in Japan: a systematic review.

Nomura S, Hisashige A, Yoneoka D, Kanda M, Miyamoto K, Segawa M, Ota E, Shibuya K - EJNMMI Res (2015)

Bottom Line: We compared the quality of studies indexed in MEDLINE with non-indexed studies, followed by a comparison of the studies' conclusions with those of international health technology assessment (HTA) reports.A significant difference was observed in several quality items between MEDLINE-indexed and non-indexed studies, although there was no difference in total quality score.Three variables (i.e., target diseases, publication year, and study type) were identified as factors related to the quality of the studies.

View Article: PubMed Central - PubMed

Affiliation: Department of Global Health Policy, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033 Japan ; Present affiliation: Department of Epidemiology and Biostatistics, School of Public Health, Imperial College London, Norfolk Place, London, W2 1PG UK.

ABSTRACT

Background: Systematic evaluations of the diagnostic accuracy of positron emission tomography (PET) imaging have been widely conducted in many countries. Although Japan's total number of PET units is the second highest in the world, very limited effort has been made to systematically assess the methodological quality of PET studies in Japan. We performed a systematic review to assess the characteristics and quality of PET diagnostic accuracy studies conducted in Japan and to analyze the factors related to their quality.

Methods: All PET studies conducted in Japan were identified using MEDLINE and the Japan Medical Abstract Society Database. The characteristics of the Japanese studies were examined and their methodological quality evaluated by the standardized quality assessment of diagnostic accuracy studies (QUADAS) tool. We compared the quality of studies indexed in MEDLINE with non-indexed studies, followed by a comparison of the studies' conclusions with those of international health technology assessment (HTA) reports.

Results: A total of 138 studies were identified. Half of them were not indexed in MEDLINE. The mean quality score of the Japanese studies was 6.7 and the proportion of high-quality studies (with a quality score higher than 8) was 32.6%. A significant difference was observed in several quality items between MEDLINE-indexed and non-indexed studies, although there was no difference in total quality score. Three variables (i.e., target diseases, publication year, and study type) were identified as factors related to the quality of the studies. Conclusions of Japanese studies relating to several target diseases were relatively consistent with international assessments.

Conclusions: Although a considerable number of diagnostic accuracy studies of PET have been conducted in Japan, a substantial proportion of high-quality studies were not indexed in international databases. High-quality Japanese studies, therefore, should be searched using Japanese databases and assessed by systematic reviews and HTA conducted internationally.

No MeSH data available.


Comparison of quality between MEDLINE-indexed and non-indexed studies.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4385095&req=5

Fig3: Comparison of quality between MEDLINE-indexed and non-indexed studies.

Mentions: Figure 3 compares the results in quality between the MEDLINE-indexed and non-indexed studies. The proportion of items categorized as ‘Yes’ among MEDLINE-indexed studies was significantly higher than among non-indexed studies in disease progression, (item 4) (p = 0.009, 30.4% for indexed studies and 8.7% for non-indexed studies); description of index test execution (item 8) (p < 0.001, 97.1% for indexed studies and 72.5% for non-indexed studies); and index test review (item 10) (p < 0.001, 75.4% for indexed studies and 43.5% for non-indexed studies). In contrast, we observed a lower proportion of ‘Yes’ items in MEDLINE-indexed studies for the description of reference test (item 9) (p < 0.001, 21.7% for indexed studies and 49.3% for non-indexed studies). However, there is no statistical difference in the mean quality score between them (p = 0.3). Note that as the language type of the paper closely corresponded to whether or not the study was indexed in MEDLINE, the differences in methodological quality between the English and Japanese papers were similar to those observed between papers indexed in MEDLINE and those that were not (data are not shown).Figure 3


An assessment of the characteristics and quality of diagnostic accuracy studies for positron emission tomography conducted in Japan: a systematic review.

Nomura S, Hisashige A, Yoneoka D, Kanda M, Miyamoto K, Segawa M, Ota E, Shibuya K - EJNMMI Res (2015)

Comparison of quality between MEDLINE-indexed and non-indexed studies.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4385095&req=5

Fig3: Comparison of quality between MEDLINE-indexed and non-indexed studies.
Mentions: Figure 3 compares the results in quality between the MEDLINE-indexed and non-indexed studies. The proportion of items categorized as ‘Yes’ among MEDLINE-indexed studies was significantly higher than among non-indexed studies in disease progression, (item 4) (p = 0.009, 30.4% for indexed studies and 8.7% for non-indexed studies); description of index test execution (item 8) (p < 0.001, 97.1% for indexed studies and 72.5% for non-indexed studies); and index test review (item 10) (p < 0.001, 75.4% for indexed studies and 43.5% for non-indexed studies). In contrast, we observed a lower proportion of ‘Yes’ items in MEDLINE-indexed studies for the description of reference test (item 9) (p < 0.001, 21.7% for indexed studies and 49.3% for non-indexed studies). However, there is no statistical difference in the mean quality score between them (p = 0.3). Note that as the language type of the paper closely corresponded to whether or not the study was indexed in MEDLINE, the differences in methodological quality between the English and Japanese papers were similar to those observed between papers indexed in MEDLINE and those that were not (data are not shown).Figure 3

Bottom Line: We compared the quality of studies indexed in MEDLINE with non-indexed studies, followed by a comparison of the studies' conclusions with those of international health technology assessment (HTA) reports.A significant difference was observed in several quality items between MEDLINE-indexed and non-indexed studies, although there was no difference in total quality score.Three variables (i.e., target diseases, publication year, and study type) were identified as factors related to the quality of the studies.

View Article: PubMed Central - PubMed

Affiliation: Department of Global Health Policy, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033 Japan ; Present affiliation: Department of Epidemiology and Biostatistics, School of Public Health, Imperial College London, Norfolk Place, London, W2 1PG UK.

ABSTRACT

Background: Systematic evaluations of the diagnostic accuracy of positron emission tomography (PET) imaging have been widely conducted in many countries. Although Japan's total number of PET units is the second highest in the world, very limited effort has been made to systematically assess the methodological quality of PET studies in Japan. We performed a systematic review to assess the characteristics and quality of PET diagnostic accuracy studies conducted in Japan and to analyze the factors related to their quality.

Methods: All PET studies conducted in Japan were identified using MEDLINE and the Japan Medical Abstract Society Database. The characteristics of the Japanese studies were examined and their methodological quality evaluated by the standardized quality assessment of diagnostic accuracy studies (QUADAS) tool. We compared the quality of studies indexed in MEDLINE with non-indexed studies, followed by a comparison of the studies' conclusions with those of international health technology assessment (HTA) reports.

Results: A total of 138 studies were identified. Half of them were not indexed in MEDLINE. The mean quality score of the Japanese studies was 6.7 and the proportion of high-quality studies (with a quality score higher than 8) was 32.6%. A significant difference was observed in several quality items between MEDLINE-indexed and non-indexed studies, although there was no difference in total quality score. Three variables (i.e., target diseases, publication year, and study type) were identified as factors related to the quality of the studies. Conclusions of Japanese studies relating to several target diseases were relatively consistent with international assessments.

Conclusions: Although a considerable number of diagnostic accuracy studies of PET have been conducted in Japan, a substantial proportion of high-quality studies were not indexed in international databases. High-quality Japanese studies, therefore, should be searched using Japanese databases and assessed by systematic reviews and HTA conducted internationally.

No MeSH data available.