Reasons and factors associated with inconclusiveness of systematic reviews about interventions for neuropathic pain
Publication: Journal of Comparative Effectiveness Research
Abstract
Aim: Systematic reviews (SRs) are frequently inconclusive. The aim of this study was to analyze factors associated with conclusiveness of SRs about efficacy and safety of interventions for neuropathic pain (NeuP). Materials & methods: The study protocol was registered in the PROSPERO database (No. CRD42015025831). Five electronic databases (Medical Literature Analysis and Retrieval System Online, Cochrane Database of Systematic Reviews, Cumulative Index for Nursing and Allied Health Literature, Database of Abstracts of Reviews of Effects and Psychological Information Database) were searched until July 2018 for SRs about NeuP management. Conclusion statements for efficacy and safety, and characteristics of SRs were analyzed. Conclusiveness was defined as explicit statement by the SR authors that one intervention is better/similar to the other in terms of efficacy and safety. Methodological quality of SRs was assessed with the AMSTAR (A MeaSurement Tool to Assess systematic Reviews) tool. Results: Of 160 SRs, 37 (23%) were conclusive for efficacy and/or safety. In the SRs, conclusions about safety were missing in half of the analyzed abstracts, and a third of the full texts. Conclusive SRs included significantly more trials and participants, searched more databases, had more authors, conducted meta-analysis, analyzed quality of evidence, and had lower methodological quality than inconclusive SRs. The most common reasons for the lack of conclusiveness indicated by the SR authors were the small number of participants and trials, and the high heterogeneity of included studies. Conclusion: Most SRs about NeuP treatment were inconclusive. Sources of inconclusiveness of NeuP reviews need to be further studied, and SR authors need to provide conclusions about both safety and efficacy of interventions.
Lay abstract
Systematic reviews (SRs) analyze the evidence from research studies to provide recommendations on the efficacy and safety of medical interventions. Clear conclusions whether an intervention is efficacious and safe are important for patients, clinicians and policymakers, to enable informed decisions. Neuropathic pain (NeuP) is caused by the injury to the nerves. In this study factors associated with conclusiveness of SRs about efficacy and safety of interventions for the treatment of NeuP were analyzed. Our results showed that the majority of analyzed SRs do not provide a definitive answer, more so if reviews were conducted by higher methodological standards. Conclusive reviews included significantly more trials and patients. Sources of inconclusiveness of NeuP reviews need to be further studied.
Systematic reviews (SRs) are frequently inconclusive, which is the usual reason why more randomized controlled trials (RCTs) or other types of primary studies are suggested in order to reach a definitive conclusion. Several research groups analyzed the conclusiveness of Cochrane SRs from various areas of research such as physiotherapy [1], pediatric gastroenterology [2], nutrition [3], neonatal themes [4], gynecological malignancy [5] and palliative care of cancer [6]. However, studies that involve both Cochrane and non-Cochrane SRs are needed, to explore reasons for inconclusiveness and factors associated with it.
In this study, the conclusiveness of SRs about interventions using neuropathic pain (NeuP) as the case study was analyzed. Pain relief is considered a basic human right [7]. Therefore, effective and safe pain therapies are of high public health significance. However, it is known that certain types of pain, such as NeuP, are still difficult to treat [8]. Hence, it is important to have conclusive evidence about the efficacy and safety of interventions for NeuP. The following main hypotheses regarding SRs about interventions for NeuP were tested: the majority of the SRs are inconclusive; conclusive SRs have included more participants, more studies, they are more commonly non-Cochrane SRs and they have higher methodological quality compared with inconclusive ones; the most common reason for inconclusiveness is the low number of participants; and the majority of conclusions presented in the SR abstract are concordant with conclusions presented in the full text.
Materials & methods
Study design
This was a methodological study based on a systematic search of the literature. Secondary data analysis was conducted only; no patient data were included in the study. The study was registered in the PROSPERO database (No. CRD42015025831).
Bibliographic search
Five electronic databases were searched: Cumulative Index for Nursing and Allied Health Literature (CINAHL), Cochrane Database of Systematic Reviews (CDSR), Database of Abstracts of Reviews of Effects (DARE), Medical Literature Analysis and Retrieval System Online (MEDLINE) and Psychological Information Database (PsycINFO) from the database inception to 17 July 2018. A combination of keywords for NeuP and systematic review/meta-analysis (search strategy was reported elsewhere [9]) was used, without language restrictions. Two authors (S Dosenovic and A Jelicic Kadic) independently conducted each of the following steps: screening of bibliographic records retrieved by search (titles and abstracts) and screening of potentially eligible full-text manuscripts. Discrepancies in judgments were resolved via discussion or involvement of a third author (L Puljak).
Inclusion criteria
SRs of RCTs on efficacy and safety of interventions for the treatment of NeuP that were published as a full text in a peer-reviewed journal were included. The definition of NeuP by the International Association for the Study of Pain (IASP) was used, which postulates that “Pain caused by a lesion or disease of the somatosensory nervous system.”, and all the accompanying notes regarding this definition were also taken into account [10]. SRs with participants of any age and gender were eligible if the participants suffered from NeuP according to the IASP definition [9]. SRs about any type of therapeutic intervention and any type of comparator were included. In terms of outcomes, SRs of RCTs that analyzed pain intensity, with or without other outcome measures were eligible for inclusion. SRs published in any language and year were included. Articles were not excluded based on the methods used; all eligible articles that were self-identified as SRs were included.
Exclusion criteria
SRs of nonrandomized studies; SRs published as conference abstracts or gray literature; and SRs of diagnostic accuracy, prognosis, prevention, epidemiology, cost–effectiveness and guidelines were excluded.
SRs with participants presenting with disorders that did not satisfy the IASP criteria, for example, participants suffering from fibromyalgia, complex regional pain syndrome type 1, low back pain without radicular pain and atypical facial pain, were excluded. Also, SRs where participants suffered from painful conditions that may be attributed to both neuropathic and nociceptive mechanisms (i.e., pain caused by multiple sclerosis, pain related to cancer and pain associated with stroke) were excluded, unless study authors categorized it as neuropathic. SRs with trials that included participants suffering from a combination of NeuP and noneligible conditions and SRs in which pain was assessed only as a composite score were also excluded.
Data extraction & categorization
The following data were extracted: year of publication, number of authors and databases searched, number of included studies, number of included patients, if the quality or risk of bias of included RCTs was assessed (yes/no), whether a meta-analysis was performed (yes/no), if the quality of body of evidence in SRs was graded (yes/no), referral to the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT) [11] guideline for selecting outcomes (yes/no), and whether it was Cochrane or a non-Cochrane SR (yes/no). The information about conclusiveness about efficacy and safety of the analyzed interventions presented in the abstract conclusion, as well as in the conclusion from the full text, was also collected. SRs were classified as conclusive (the authors explicitly stated that one intervention is better than the other, or one intervention is similar to the other in terms of efficacy and safety) or inconclusive (the SR authors indicated there were insufficient data, the available data are of low or inadequate quality or that further research is needed to make conclusions). If either efficacy or safety conclusion statement from the abstract or the full text was inconclusive, the whole SR was categorized as inconclusive. If both efficacy and safety were conclusive or efficacy was conclusive, and a safety conclusion was not reported, the SR was categorized as conclusive. For SRs that investigated multiple interventions, if at least one intervention was found to be inconclusive, the whole SR was rated as such.
For inconclusive reviews reasons for the lack of conclusiveness were extracted and categorized. Data extractions were done independently in duplicate (I Vuka and D Nujic; G Tintor and A Dujmic) and discrepancies were resolved by another author (S Dosenovic).
Methodological quality appraisal
The methodological quality of SRs was analyzed with the AMSTAR (A MeaSurement Tool to Assess systematic Reviews) checklist [12]. Before the quality assessment, the items on both scales were discussed, and a calibration exercise was performed on ten included SRs. Two authors independently rated each SR (I Vuka, D Nujic) and discrepancies were resolved by the involvement of a third author (S Dosenovic).
Analyses
The year of publication, number of studies and patients, number of authors and databases searched; the number of SRs which assessed the quality or risk of bias of RCTs, performed a meta-analysis, graded the quality of evidence, and referred to IMMPACT guideline for selecting outcomes; the frequency of Cochrane and non-Cochrane reviews, and methodological quality of SRs in conclusive versus inconclusive reviews were compared. Concordance between conclusions presented in the abstracts of SRs and conclusions presented in the full texts was also analyzed. ‘Empty reviews’, in other words, SRs without included RCTs were analyzed separately.
The data were presented as frequencies, percentages, and median and interquartile range (IQR). Normality distribution of data was tested with the Kolmogorov–Smirnov test. The Mann–Whitney test was used for differences in continuous data because data were not normally distributed, and chi-square test was used for differences in proportions. IBM SPSS Statistics for Windows software, version 19.0.0 (IBM Corp., NY, USA), was used for statistical analyses. Statistical significance was defined at p < 0.05, two sided.
Results
After screening titles and abstracts of 4644 items retrieved from the five bibliographic databases, 3824 items were excluded. Subsequently, 468 full-text manuscripts were retrieved for detailed eligibility assessment. Further 300 manuscripts were excluded because they did not fit inclusion criteria. Thus, 168 SRs of RCTs about interventions for NeuP were found. However, of those 168, there were eight empty reviews. They were analyzed separately. Therefore, 160 SRs were included in the main analysis. The characteristics of those SRs are presented in Supplementary Table 1.
Of the 160 SRs with included RCTs, 37 (23%) had conclusive statements about efficacy and/or safety (Table 1). All of the 160 SRs contained conclusions about efficacy, but 93 (53%) SRs did not provide any conclusions about the safety of the analyzed intervention(s) in the abstract; while that number was lower in the full texts (n = 62; 34%).
| Variable | All SRs (n = 160) | Conclusive SRs for efficacy and/or safety (n = 37) | Inconclusive SRs (N = 123) | p-value |
|---|---|---|---|---|
| Patients enrolled (n), median (IQR) | 817 (353–1817.5) | 1596 (478–2938.5) | 728 (317–1410) | U = 1146, p < 0.001† |
| RCTs included (n), median (IQR) | 10 (5–17) | 12 (7–24) | 9 (5–16) | U = 1630.5, p = 0.009† |
| Authors (n), median (IQR) | 5 (3–6.75) | 6 (4–7.5) | 5 (3–6) | U = 1739, p = 0.028† |
| Databases searched (n), median (IQR) | 4 (3–6) | 4 (3–5) | 5 (3–6) | U = 2879.5, p = 0.013† |
| Reference to IMMPACT guideline for selecting outcomes, n (%) | 32 (20) | 5 (14) | 26 (21) | X2 = 1.620, p = 0.445 |
| Meta-analysis performed, n (%) | 122 (76) | 36 (97) | 86 (70) | X2 = 11.786, p = 0.003† |
| Quality or RoB of RCTs assessed, n (%) | 151 (94) | 33 (89) | 118 (96) | X2 = 3.688, p = 0.158 |
| Quality of evidence in SRs assessed, n (%) | 60 (38) | 8 (22) | 52 (42) | X2 = 6.143, p = 0.046† |
| Year of SR publication, median (IQR) | 2014 (2011–2016) | 2014 (2010.5–2016) | 2014 (2011–2016) | U = 2274, p = 0.997 |
| Cochrane reviews, n (%) | 44 (28) | 4 (11) | 40 (33) | X2 = 6.724, p = 0.010† |
| Methodological quality, AMSTAR score, median (IQR) | 7 (6–9) | 6 (5–7.5) | 8 (6–10) | U = 3125, p < 0.001† |
†
Statistically significant results.
X2 = Chi square test; AMSTAR: A MeaSurement Tool to Assess systematic Reviews; IMMPACT: Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials; IQR: Interquartile range; RCT: Randomized controlled trial; RoB: Risk of Bias; SR: Systematic review; U: Mann–Whitney U-test.
Factors associated with conclusiveness
SRs that were conclusive for efficacy and/or safety enrolled more patients compared with inconclusive ones (median, IQR: 1596 [478–2938.5] vs 728 [317–1410], p < 0.001). Likewise, SRs with conclusive evidence included significantly more RCTs (median, IQR: 12 [7–24] vs 9 [5–16], p = 0.009). There was also a significant association between conclusiveness of SRs and performed meta-analysis (X2 = 11.786, p = 0.003), and the quality level of a body of evidence assessment (X2 = 6.143, p = 0.046).
Inconclusive SRs had significantly higher methodological quality than conclusive SRs on the AMSTAR checklist (median, IQR: 8 [6–10] vs 6 [5–7.5], p < 0.001). Significantly higher number of Cochrane SRs was found among inconclusive SRs (n = 40; 33%), compared with conclusive SRs (n = 4; 11%; p = 0.010).
Other statistically significant factors associated with conclusiveness were the number of authors and the number of databases searched (Table 1).
Publication date, analyzed quality or risk of bias of RCTs, and the reference to IMMPACT guideline for selecting relevant outcomes in chronic pain trials were not associated with conclusiveness (Table 1).
Reasons for inconclusiveness
SR authors indicated 27 different reasons for the lack of conclusiveness. The six most common reasons were: small number of included participants; small number of included studies; high heterogeneity of included studies, such as different doses and different duration of treatment, heterogeneous population, etc.; insufficient information about methodology or poor methodology; missing, inconsistent or partial data about safety; and low quality of evidence (Table 2). Conclusion statements about the efficacy and safety of analyzed NeuP interventions and categorizations are shown in Supplementary Table 2.
| Reason | n (%) of SRs |
|---|---|
| Small number of participants | 63 (51) |
| Small number of RCTs | 51 (41) |
| High heterogeneity of included studies (different interventions, methodologies and population) | 43 (35) |
| Insufficient information about methodology or poor methodology | 43 (35) |
| Missing, inconsistent or partial data about safety; low number of events | 42 (34) |
| Low/very low quality of evidence | 38 (31) |
| Lack of long-term outcome assessment/insufficient information | 36 (29) |
| Inadequacy of outcome measures/lack of relevant outcomes | 28 (23) |
| High risk of bias for at least one domain | 27 (22) |
| Heterogeneity of outcome measures | 19 (15) |
| Selection bias: issues with randomization and allocation concealment | 17 (14) |
| Performance bias: issues with blinding of participants and personnel | 16 (13) |
| Lack of adequate comparisons | 15 (12) |
| Attrition bias: lack of participant follow-up or inadequate follow-up, attrition not reported and high withdrawal rate of participants | 13 (11) |
| Detection bias: problems with blinding of outcome assessors | 12 (10) |
| Publication bias, language bias, relevant trials potentially missed | 10 (8) |
| Cross-over trials, wash-out period in cross-over trial inadequate or not reported | 6 (5) |
| Reporting bias: selective reporting of outcomes | 6 (5) |
| Insufficient data; lack of available literature | 5 (4) |
| Lack of clear diagnostic criteria/inconsistent diagnostic criteria | 4 (3) |
| Large placebo responses | 3 (2) |
| Geographically limited area, no ethnic diversity | 2 (2) |
| Unclear inclusion and exclusion criteria | 2 (2) |
| Commercial bias | 2 (2) |
| Limited external validity and generalizability | 2 (2) |
| Insufficient information about quality of drugs/interventions | 1 (1) |
| Lack of direct evidence about efficacy, only surrogate outcomes | 1 (1) |
RCT: Randomized controlled trial; SR: Systematic review.
Concordance between conclusions in the abstract & main text of SRs
Concordance between 155 (97%) SR abstract and full-text conclusions was found. Only five pairs of abstracts/full texts were discordant in efficacy conclusion [13–17]. In all five cases the abstract was conclusive, and in the full text the authors indicated inconclusiveness.
Empty reviews
There were eight SRs without included RCTs (Supplementary Table 2). As such, they were inconclusive for both efficacy and safety. All of them were Cochrane reviews.
Discussion
The analysis of 160 SRs about NeuP indicated that the majority of SRs had inconclusive statements regarding results about the efficacy and/or safety. A small number of included participants was the most commonly stated reason by the SR authors to explain inconclusive results. Conclusiveness of evidence was associated with SRs that included more participants, more RCTs, searched more databases, had more authors, conducted meta-analysis and analyzed the quality of evidence. On the other hand, inconclusive SRs had a higher methodological quality.
The conclusiveness of results in SRs is important for patients, clinicians and policymakers, to enable informed decisions and save time when searching for definitive information. The results of this study indicated that the majority of analyzed SRs in the field of NeuP do not provide a definitive answer, even more frequently if reviews were published by Cochrane.
In SRs about NeuP, a higher number of included participants and studies were found to be associated with conclusiveness. Larger trials with many patients will contribute more toward the estimate of the magnitude of effect [18], and subsequently to the more conclusive results in evidence syntheses. The most common reasons for inconclusiveness that the SR authors provided were the low number of participants, stated in half of SRs, followed by a small number of included RCTs, stated by 41% of SRs.
Heterogeneity of the included studies, as a reason for inconclusiveness, is another important cause for concern. Nonstandardized interventions, methodologies and population were stated as the third most common reason for the lack of conclusiveness, together with issues related to the methodological quality of RCTs, each stated by a third of SR authors. Arguments about including primary studies of low quality and with a low number of patients in SRs are ongoing, with valid arguments on both sides of the discussion [19,20].
Inconclusive SRs had significantly higher methodological quality compared with the conclusive ones; the difference was 2 out of 11 AMSTAR points. This finding might indicate that authors of higher-quality reviews may be more critical toward evidence they are analyzing.
Another finding of this study is that inconclusive SRs more frequently rated the quality of evidence in a systematic and transparent manner, such as using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach [21]. In line with this finding, low/very low quality of evidence was frequently stated as a cause for inconclusiveness by authors of SRs in this sample. However, it also needs to be emphasized that the focus of this study was analysis of conclusiveness as explicitly reported by SR authors at the end of the abstract and full text; past results indicate that authors may report conclusive statements about effects of an intervention, despite having a GRADE assessment in which primary outcome was rated less than high quality [22].
SRs should be used as a guide for justifying the conduct of new clinical trials, otherwise, a trial may represent waste in research [23]. The current study indicated a high prevalence of inconclusiveness of available SRs in the field of NeuP, which would imply a need for conducting many new trials in this field. Some of the hypotheses, which were based on the prior research findings in the field, were confirmed, while others were not. These previous results were reported in a number of methodological studies that analyzed the conclusiveness of Cochrane reviews in various research fields [1–6,24] and most of those studies, except two [1,6], found that the majority of analyzed Cochrane SRs had conclusive evidence. A high prevalence of inconclusive Cochrane reviews was observed in a study by Momosaki et al., which found 94.3% of physiotherapy Cochrane reviews to be inconclusive [1]. Zhang et al. showed that less than half of Cochrane reviews in palliative and supportive care for cancer were conclusive [6]. Conclusiveness of SRs varied with different research areas.
A difference in the proportion of Cochrane and non-Cochrane reviews among the SRs with conclusive evidence was found, in other words, nine in ten Cochrane reviews about NeuP treatment were inconclusive. A part of the reason for this finding might be that Cochrane reviews in various fields [25–29], including NeuP [30], were found to have higher methodological quality.
Appropriate outcomes should be selected while designing clinical trials in order to allow the direct comparison of the analyzed interventions. Measuring and reporting relevant outcomes is one of the fundamental features of comparative effectiveness research [31]. The standardized core outcome set and outcome measures are selected for specific research areas to reduce this heterogeneity [32]. Heterogeneity of outcomes was stated as one of the reasons behind inconclusiveness by 15% of SRs in the sample. Several studies have previously shown that recommended core outcome set and core measures are insufficiently utilized in pain research [33–37]. In the analyzed sample there was no significant difference in the conclusiveness of SRs based on the date of publication, analyzed quality or risk of bias of included RCTs and the reference to IMMPACT guideline for selecting relevant core outcomes in chronic pain trials [11].
It is worrying that every other SR did not have a conclusion about the safety of analyzed interventions in the abstract. Readers of scientific literature often rely only on abstracts to get quick information whether an intervention works and is it safe, so this is a limitation of the published literature. Furthermore, a third of the full-text manuscripts did not contain a safety conclusion. Safety outcomes were frequently missing from RCTs; a third of SR authors indicated that data about safety outcomes were missing, inconsistent or partial.
This study, although based on a large sample of SRs that were systematically searched without language restrictions, had several limitations. First, unpublished (gray) literature was not included. Second, the search was limited to SRs published by mid-2018; however, the purpose of the study was not to summarize the most current literature about the treatment of this chronic condition, but to look behind reasons for the lack of conclusive evidence. Additionally, the association between the reporting quality of SRs, using a tool such as the Preferred reporting items for systematic reviews and meta-analyses checklist [38], and conclusiveness of NeuP SRs was not assessed. It would be interesting to investigate whether there is any association between reporting quality and statements about conclusiveness of SRs.
In our study we have depended on the authors’ reporting of conclusiveness, without going deeply into the definition of conclusiveness. Currently, there are no consensus guidelines regarding the criteria that would define a conclusive review, and few studies have addressed that issue. This has important implications on decisions such as whether a SR needs an update and whether new primary studies are needed, and potentially preventing research waste. Cochrane, for example, has labeled some of their reviews as ‘stable’, meaning that they are not in a need of updating, but it has been shown recently that justifications for stabilizing a Cochrane review were often not clear and transparent [39]. Thus, further studies regarding conclusiveness of SRs are needed.
Conclusion
In conclusion, the insufficient number of participants and studies are among the most commonly indicated reasons for the inconclusiveness of results in SRs about interventions for NeuP. The performed analyses confirmed the association of conclusiveness with the number of included participants and RCTs, as well as with the methodological quality of SRs, evidence synthesis and rated quality of evidence by SR authors, the number of searched databases, and the number of authors. Sources of inconclusiveness of SRs need to be further studied and authors of SRs need to provide conclusions about both efficacy and safety of analyzed interventions.
•
This study analyzed factors associated with conclusiveness of systematic reviews (SRs) about efficacy and safety of interventions for neuropathic pain (NeuP).
•
Five electronic databases were searched for SRs about NeuP management.
•
Of 160 included SRs, only 37 (23%) were conclusive for efficacy and/or safety.
•
A small number of included participants was the most commonly stated reason by the SR authors to explain inconclusive results.
•
Conclusiveness of evidence was associated with SRs that included more participants, more randomized controlled trials, searched more databases, had more authors, conducted meta-analysis and analyzed the quality of evidence. Inconclusive SRs had a higher methodological quality.
•
Every other SR did not have a conclusion about the safety of analyzed interventions in the abstract, as well as a third of the full-text manuscripts.
•
A high prevalence of inconclusiveness of available NeuP SRs would imply a need for conducting many new trials in this field.
•
Authors of SRs need to provide conclusions about both efficacy and safety of analyzed interventions.
Author contributions
S Dosenovic and L Puljak provided substantial contributions to the conception or design of the work, analysis and interpretation of data for the work, and drafting of the manuscript. S Dosenovic, A Dujmic, D Nujic, I Vuka, G Tintor and A Jelicic Kadic contributed to the acquisition of data for the work. All authors contributed to the revision of manuscript critically for important intellectual content, final approval of the version to be published and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Acknowledgments
The authors are grateful to I Ramić for her help with the test screening of titles and abstracts.
Financial & competing interests disclosure
The study was conducted within the project ‘Treating neuropathic pain with dorsal root ganglion stimulation’ (HRZZ-IP-2013-11-4126) awarded to D Sapunar by the Croatian Science Foundation (HRZZ). Even though the study was conducted within the project, the study did not receive any funding from the project. I Vuka’s salary was funded by: the grant of the Croatian Science Foundation (Hrvatska zaklada za znanost, HRZZ) while working on this manuscript; grant for Young Scientist Career Development (HRZZ-DOK-2015-10-2774). The HRZZ was not involved in any way in the design of the study, collection, analysis, interpretation of data or writing the manuscript. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
No writing assistance was utilized in the production of this manuscript.
Supplementary Material
References
1.
Momosaki R, Tsuboi M, Yasufuku Y et al. Conclusiveness of Cochrane reviews in physiotherapy: a systematic search and analytical review. Int. J. Rehabil. Res. 42(2), 97–105 (2019).
2.
Cohen S, Lubetzky R, Mimouni FB, Marom R, Mandel D. Conclusiveness of the Cochrane reviews in pediatric-gastroenterology: a systematic analysis. Eur. J. Gastroenterol. Hepatol. 25(2), 252–254 (2013).
3.
Cohen S, Mandel D, Mimouni FB, Marom R, Lubetzky R. Conclusiveness of the Cochrane reviews in nutrition: a systematic analysis. Eur. J. Clin. Nutr. 68(2), 143–145 (2014).
4.
Mandel D, Littner Y, Mimouni FB, Lubetzky R. Conclusiveness of the Cochrane neonatal reviews: a systematic analysis. Acta Paediatr. 95(10), 1209–1212 (2006).
5.
Yin S, Chuai Y, Wang A, Zhang L. Conclusiveness of the Cochrane reviews in gynaecological cancer: a systematic analysis. J. Int. Med. Res. 43(3), 311–315 (2015).
6.
Zhang X, Wu Z, Zhao H, Li C, Wu J, Dai G. Conclusiveness of the Cochrane reviews in palliative and supportive care for cancer: a systematic analysis. Am. J. Hosp. Palliat. Med. 34(1), 53–56 (2017).
7.
Lohman D, Schleifer R, Amon JJ. Access to pain treatment as a human right. BMC Med. 8(1), 8 (2010).
8.
Dworkin RH, O'connor AB, Backonja M et al. Pharmacologic management of neuropathic pain: evidence-based recommendations. Pain 132(3), 237–251 (2007).
9.
Dosenovic S, Jelicic Kadic A, Miljanovic M et al. Interventions for neuropathic pain: an overview of systematic reviews. Anesth. Analg. 125(2), 643–652 (2017).
10.
IASP. IASP taxonomy (2012). www.iasp-pain.org/terminology?navItemNumber=576#Neuropathicpain
11.
Turk DC, Dworkin RH, Allen RR et al. Core outcome domains for chronic pain clinical trials: IMMPACT recommendations. Pain 106(3), 337–345 (2003).
12.
Shea BJ, Grimshaw JM, Wells GA et al. Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Med. Res. Methodol. 7, 10 (2007).
13.
Stein C, Eibel B, Sbruzzi G, Lago PD, Plentz RDM. Electrical stimulation and electromagnetic field use in patients with diabetic neuropathy: systematic review and meta-analysis. Braz. J. Phys. Ther. 17(2), 93–104 (2013).
14.
Zhang WW, Li MQ, Liu L. [Meta-analysis of gabapentin in the treatment of postherpetic neuralgia]. Chin. J. Contemp. Neurol. Neurosurg. 13(9), 760–765 (2013).
15.
Watts RW, Silagy CA. A meta-analysis on the efficacy of epidural corticosteroids in the treatment of sciatica. Anaesth. Intensive Care 23(5), 564–569 (1995).
16.
Zheng C, Ou W, Shen H, Zhou Z, Wang J. Combined therapy of diabetic peripheral neuropathy with breviscapine and mecobalamin: a systematic review and a meta-analysis of Chinese studies. Biomed Res. Int. 2015, 680756 (2015).
17.
Wang SL, Wang H, Nie HY, Bu G, Shen XD. The efficacy of pregabalin for acute pain control in herpetic neuralgia patients: a meta-analysis. Medicine (Baltimore) 96(51), e9167 (2017).
18.
Guyatt GH, Oxman AD, Vist G et al. GRADE guidelines: 4. Rating the quality of evidence--study limitations (risk of bias). J. Clin. Epidemiol. 64(4), 407–415 (2011).
19.
Handoll HHG, Langhorne P. In defence of reviews of small trials: underpinning the generation of evidence to inform practice. Cochrane Database Syst Rev. 11 (2015).
20.
Roberts I, Ker K. How systematic reviews cause research waste. Lancet 386(10003), 1536 (2015).
21.
Guyatt G, Oxman AD, Akl EA et al. GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. J. Clin. Epidemiol. 64(4), 383–394 (2011).
22.
Conway A, Conway Z, Soalheira K, Sutherland J. High quality of evidence is uncommon in Cochrane systematic reviews in anaesthesia, critical care and emergency medicine. Eur. J. Anaesthesiol. 34(12), 808–813 (2017).
23.
Cooper NJ, Jones DR, Sutton AJ. The use of systematic reviews when designing studies. Clin. Trials 2(3), 260–264 (2005).
24.
Mimouni M, Mimouni F, Segev F. Conclusiveness of the Cochrane eye and vision group reviews. BMC Res. Notes 8(1), 242 (2015).
25.
Moseley AM, Elkins MR, Herbert RD, Maher CG, Sherrington C. Cochrane reviews used more rigorous methods than non-Cochrane reviews: survey of systematic reviews in physiotherapy. J. Clin. Epidemiol. 62(10), 1021–1030 (2009).
26.
Fleming PS, Seehra J, Polychronopoulou A, Fedorowicz Z, Pandis N. Cochrane and non-Cochrane systematic reviews in leading orthodontic journals: a quality paradigm? Eur. J. Orthod. 35(2), 244–248 (2013).
27.
Petticrew M, Wilson P, Wright K, Song F. Quality of Cochrane reviews is better than that of non-Cochrane reviews. BMJ 324(7336), 545 (2002).
28.
Shea BJ, Hamel C, Wells GA et al. AMSTAR is a reliable and valid measurement tool to assess the methodological quality of systematic reviews. J. Clin. Epidemiol. 62(10), 1013–1020 (2009).
29.
Moher D, Tetzlaff J, Tricco AC, Sampson M, Altman DG. Epidemiology and reporting characteristics of systematic reviews. PLoS Med. 4(3), e78 (2007).
30.
Dosenovic S, Jelicic Kadic A, Vucic K, Markovina N, Pieper D, Puljak L. Comparison of methodological quality rating of systematic reviews on neuropathic pain using AMSTAR and R-AMSTAR. BMC Med. Res. Methodol. 18(1), 37 (2018).
31.
Tunis SR, Clarke M, Gorst SL et al. Improving the relevance and consistency of outcomes in comparative effectiveness research. J. Comp. Eff. Res. 5(2), 193–205 (2016).
32.
Williamson PR, Altman DG, Blazeby JM et al. Developing core outcome sets for clinical trials: issues to consider. Trials 13, 132–132 (2012).
33.
Krsticevic M, Dosenovic S, Dimcea DA et al. Outcome domains, outcome measures, and characteristics of randomized controlled trials testing nonsurgical interventions for osteoarthritis. J. Rheumatol. 47(1), 126–131 (2020).
34.
Puljak L, Dosenovic S, Boric K. Importance of consistent outcomes in randomized controlled trials and systematic reviews about anesthesiology and pain. Pain Manag. 8(4), 251–253 (2018).
35.
Boric K, Boric M, Dosenovic S et al. Authors’ lack of awareness and use of core outcome set on postoperative pain in children is hindering comparative effectiveness research. J. Comp. Eff. Res. 7(5), 463–470 (2018).
36.
Dosenovic S, Nikolic Z, Ivancev B, Kadic AJ, Puljak L. Awareness and acceptability of Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials core outcome set for chronic pain among surveyed neuropathic pain authors. J. Comp. Eff. Res. 8(9), 671–683 (2019).
37.
Dosenovic S, Jelicic Kadic A, Jeric M et al. Efficacy and safety outcome domains and outcome measures in systematic reviews of neuropathic pain conditions. Clin. J. Pain 34(7), 674–684 (2018).
38.
Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 6(7), e1000097 (2009).
39.
Babić A, Poklepovic Pericic T, Pieper D, Puljak L. How to decide whether a systematic review is stable and not in need of updating: analysis of Cochrane reviews. Res. Synth. Methods (2020) (Epub ahead of print).
Information & Authors
Information
Published In
Pages: 67 - 75
PubMed: 33355481
Copyright
© 2020 Future Medicine Ltd.
History
Received: 6 August 2020
Accepted: 28 October 2020
Published online: 23 December 2020
Keywords:
Topics
Authors
Metrics & Citations
Metrics
Article Usage
Article usage data only available from February 2023. Historical article usage data, showing the number of article downloads, is available upon request.
Citations
How to Cite
Reasons and factors associated with inconclusiveness of systematic reviews about interventions for neuropathic pain. (2020) Journal of Comparative Effectiveness Research. DOI: 10.2217/cer-2020-0165
Export citation
Select the citation format you wish to export for this article or chapter.