| Objective?To evaluate the methodological and reporting quality of systematic reviews and meta-analyses on nursing interventions in the field of heart failure;?To explore the influence factor of methodological and reporting quality of systematic reviews and meta-analysis of nursing intervention for patients with heart failure and,so it provides references for standardizing the reporting process,improving the methodological quality of research and enhancing the scientific quality of literature;?To analyze whether the publication of AMSTAR and PRISMA scales can improve the methodological and reporting quality of systematic reviews and meta-analyses on nursing interventions in the field of heart failure.Method?Computer queries the papers published in databases of Embase,Pubmed and Cochrane Database of Systematic Reviews for systematic reviews/meta-analyses on nursing intervention for heart failure patients published from inception of databases to October 1th 2018;?Through two authors independently selected papers and extracted data,the methodological and reporting quality of the included articles are evaluated according to 11 items in A Measurement Tool to Assess Systematic Reviews(AMSTAR)scale and 27 items in Preferred Reporting Items for Systematic Reviews and Meta Analyses(PRISMA)scale,respectively;?Variables are extracted included: name of the first author,country of the first author,whether there are co-authors of statistics and epidemiology,year of publication,number of original studies included,number of citations,name of source magazine,influencing factors of source magazine,whether the magazine is a science Citation Index(SCI),whether the magazine is a nursing magazine,whether the literature contains meta-analysis,whether it belongs to Cochrane,whether original research included Randomized Control Trial(RCT),etc;?Statistical analysis: The total scores of AMSTAR and PRISMA scales and the reporting rate of each item are calculated.T-test is used to compare the two groups of mean,and one-way ANOVA is used to compare the multiple groups of mean,where ?=0.05.Documents with PRISMA score not less than 21.00 are divided into high-quality group,which are less 21.00 divided into middle-low-quality group.High-quality predictors are analyzed by single-factor logistic regression.Variables with P<0.05 are analyzed by multi-factor logistic regression,which are selected step by step.All data are input into SPSS18.0 statistical software for analysis and calculation.Results1.The characteristics of included study:There are 69 articles in the field of nursing intervention for patients with heart failure,where 19 of the articles have statisticians and epidemiology as coauthor and 12 are written in accordance with the PRISMA statement.2.Methodological quality of included study:?The average AMSTAR score of systematic reviews and meta-analyses on nursing intervention in patients with heart failure is 7.15 ±2.81 point;?The AMSTAR items with reporting rate of more than 70.00%: item 3(comprehensive literature search,76.80%),item 5(List of studies provided,81.20%),item6(characteristics of included studies,78.30%),item 9(appropriate methods used to combine,95.70%)and item 11(conflict of interest stated,76.80%);?The AMSTAR items with reporting rate of less than 50%: item 1(a priori study design stated,29.00%),item 4(status of publication used as inclusion,47.80%)and item 10(likelihood of publication bias,27.50%);?The results of univariate analysis are showed that the average score of AMSTAR scale is higher,where P<0.05 in the literature with not less than three authors,including Meta-analysis,co-authors of statistics and epidemiology,published in Cochrane Library,pre-design and non-nursing journals;?Multivariate Logistic regression analysis are showed that the quality of literature methodology including meta-analysis is higher than that without meta-analysis,where OR = 49.60,95 CI%: 8.00~226.70,P<0.01,and the quality of literature methodology with pre-design scheme is higher than that without or without reference,where OR = 19.60,95CI%: 2.07~185.27,P = 0.003;?Covariance analysis: including Meta-analysis(MD = 2.31)and previous design(MD =1.54)would affect the AMSTAR score(P<0.05);?After the publication of AMSTAR scale,the adherence of following AMSTAR items was improved: item 7(scientific quality of included studies,77.78% vs 46.67%,OR = 4.00,95 CI%: 1.20~13.28)and item 11(conflict of interest stated,81.48% vs 46.67%,OR = 5.03,95 CI%: 1.48~17.12);?After the publication of PRISMA scale,the adherence of following AMSTAR items was improved: item 7(scientific quality of included studies,76.47% vs 44.44%,OR = 4.06,95% CI: 1.31~12.61)and item 11(conflict of interest stated,82.35% vs 50.00%,OR = 4.67,95% CI: 1.45~15.06).3.reporting quality of included study:?The average PRISMA score of systematic reviews and meta-analyses on nursing intervention in patients with heart failure is 19.86+4.86 point;?The PRISMA items with a reporting rate of more than 70.00% are as follows: item 2(structured summary,89.90%),item 3(rational,97.10%),item 4(objective,97.10%),item 6(eligibility criteria,92.80%),item 7(information sources,97.10%),item 9(study selection,75.40%),item 12(risk of bias in individual studies,72.50%),item 14(synthesis of results,94.20%),item 16(additional analyses,72.50%),item 18(study characteristics,78.30%),item 20(results of individual studies,88.40%),item 21(synthesis of results,92.80%),item 23(additional analyses,78.30%),item 24(summary of evidence,92.80%),item 25(limitations,82.60%)and item 26(conclusions,95.70%);?The PRISMA items with a reporting rate of less than 50.00% are as follows: item 15(Risk of bias across studies,24.60%)and item 22(risk of bias across studies,24.60%);?Univariate analysis are showed that the systematic reviews and meta-analysis with the characteristics of not less than three authors,including meta-analysis,co-authors of statistics and epidemiology,published in the Cochrane library,pre-design scheme,funded and included RCT in the original research are and published in non-nursing journals are higher about the average score of scale PRISMA,where P < 0.05;?Multivariate Logistic regression analysis are showed that the quality of literature reports contained meta-analysis is higher than that without meta-analysis(OR=21.24,95 CI%:4.82~93.55,P<0.001),and the quality of literature reports with pre-design schemes is higher than that without or without reference(OR=7.78,95 CI%:1.56~38.86,P=0.012);?Covariance analysis: the number of authors(MD=2.01),including meta-analysis(MD=4.01)and publication of AMSTAR(MD=2.44),affected the score of PRISMA scale(P<0.05);?After the publication of the AMSTAR scale,the adherence of following PRISMA items was improved: item 12(risk of bias in individual studies,81.48% vs 53.33%,OR = 3.85,95% CI: 1.13~13.10),item 13(summary measures,81.48% vs 53.33%,OR = 3.85,95% CI: 1.13~13.10)and item 26(conclusions,100.00% vs 80.00%,OR = 30.52,95% CI:1.48~629.32);?After the publication of the PRISMA scale,the adherence of following PRISMA items was improved: item 12(risk of bias in individual studies,82.35% vs 55.56%,OR = 3.73,95% CI: 1.15~12.10),item 13(effect index,84.31% vs 55.56%,OR = 4.30,95% CI: 1.30~14.24)and item 26(conclusion,100.00% vs 83.33%,OR = 23.26,95% CI: 1.14~475.21).Conclusion:The quality of systematic reviews and meta-analysis published in the field of nursing intervention for patients with heart failure are of medium quality,and the credibility of the evidence provided from papers is limited,so caution should be taken when applying the results.The methodological quality of systematic reviews and meta-analysis of the included documents is generally high with pre-design schemes and meta analyse,the reporting quality of systematic reviews and meta-analysis of the included documents is generally high with in the 3 and above authors and meta-analyse,evidence users can refer to such documents more.After the publication of the PRISMA and AMSTAR scales,the reporting of systematic reviews and meta-analysis have improved on the evaluation and reporting quality of included study,conflict of interest,impact indicators and conclusions.Peer review experts and editorial departments of journals should vigorously promote and apply the PRISMA statement and AMSTAR scale.Researchers should strictly follow the PRISMA statement and AMSTAR scale when making systematic reviews and meta-analysis,so as to make the research results true and scientific to serve medical and health practice. |
PDF Full Text Request