Letter to the Editor
To the Editors
There is a great need for high-quality systematic reviews that improve clinical management of bacterial vaginosis (BV), but we have serious reservations about a review by Mehta1 published in the October 2012 issue of Sexually Transmitted Diseases. Dr Mehta fails to add to the existing literature because the same 6 studies she reviews were previously evaluated in a 1999 systematic review by Potter.2 At a minimum, we would expect her to be aware of that review and give reasons why it needs updating. This omission and the lack of acknowledgement of Potter’s work are a serious oversight.
The research community has gone to great lengths since Potter’s time to improve the quality of systematic reviews through the development of the PRISMA statement.3 The methods described in the recommendations are sorely lacking in Mehta’s work. Her limited search strategy, lack of procedures for minimizing potential bias, and failure to describe review processes have produced a work of questionable value. Not acknowledging Potter’s review also raises serious questions about the validity of her search. In the future, we hope that authors like Mehta will register their systematic reviews with a Web site such as PROSPERO so that unneeded duplication of effort like this can be avoided.4 A review by a single author using a single database also has potential for compromising validity in results. Authors should use several databases such as Cochrane, EMBASE, or CENTRAL to minimize this risk.5,6 Mehta also fails to include her review protocol, thus preventing replication or future updates.7 Neither does she share her inclusion and exclusion criteria,8 provide information about the period for which data were extracted, or identify “controlled vocabulary” for indexing PubMed such as MeSH terms. Conducting a scoping search of PubMed for instance, we were able to identify a randomized clinical trial showing positive results for a treatment regimen for BV in women that included partner treatment that has not been included in the review.9 We do not know if it was excluded for some reason, or simply not identified in the original search. Furthermore, conducting a hand search of references included in a review on BV treatment,10 we found an additional 4 studies that included partner treatment.11–14 We have no idea if these studies should have been included or might have altered the conclusions of Mehta’s review. Although we are aware of the controversy surrounding the use of English-only language literature15,16 and unpublished data,17 it may also have been useful if Mehta had discussed some ineligible studies to give additional context and completeness to the analysis and possibly suggest new avenues for research. Finally, although Mehta reports that studies reviewed are heterogeneous, she fails to suggest how this might impact her analysis.
It is important for researchers conducting reviews to be thorough and rigorous to ensure that results contribute to improved policy and practice. In these respects, Mehta’s work highlights the types of shortcomings that can compromise the important value of systematic reviews.
Purnima Madhivanan, MPH, PhD
Giselle A. Barreto, BA
Amisha Revawala, BDS
Chelsie Anderson, MA
Sheila McKinney, MA
Dudith Pierre-Victor, MPH
Robert Stempel College of Public
Health and Social Work
Florida International University
1. Mehta SD. Systematic review of randomized trials of treatment of male sexual partners for improved bacteria vaginosis outcomes in women. Sex Transm Dis 2012; 39: 822–830.
2. Potter J. Should sexual partners of women with bacterial vaginosis receive treatment? Br J Gen Pract 1999; 49: 913–918.
3. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. BMJ 2009; 339: b2535.
4. Booth A, Clarke M, Ghersi D, et al. An international registry of systematic-review protocols. Lancet 2011; 377: 108–109.
5. Jadad AR, McQuay HJ. Searching the literature. Be systematic in your searching. BMJ 1993; 307: 66.
6. Lemeshow AR, Blum RE, Berlin JA, et al. Searching one or two databases was insufficient for meta-analysis of observational studies. J Clin Epidemiol 2005; 58: 867–873.
7. PLoS Medicine Editors. Best practice in systematic reviews: The importance of protocols and registration. PLoS Med 2011; 8: e1001009.
8. Dickersin K, Scherer R, Lefebvre C. Identifying relevant studies for systematic reviews. BMJ 1994; 309: 1286–1291.
9. Larsson PG, Brandsborg E, Forsum U, et al. Extended antimicrobial treatment of bacterial vaginosis combined with human lactobacilli to find the best treatment and minimize the risk of relapses. BMC Infect Dis 2011; 11: 223.
10. Koumans EH, Markowitz LE, Hogan V. Indications for therapy and treatment recommendations for bacterial vaginosis in nonpregnant and pregnant women: A synthesis of data. Clin Infect Dis 2002; 35: S152–S172.
11. Eschenbach DA, Critchlow CW, Watkins H, et al. A dose-duration study of metronidazole for the treatment of nonspecific vaginosis. Scand J Infect Dis Suppl 1983; 40: 73–80.
12. Hagstrom B, Lindstedt J. Comparison of two different regimens of metronidazole in the treatment of non-specific vaginitis. Scand J Infect Dis Suppl 1983; 40: 95–96.
13. Hovik P. Nonspecific vaginitis in an outpatient clinic. Comparison of three dosage regimens of metronidazole. Scand J Infect Dis Suppl 1983; 40: 107–110.
14. Jerve F, Berdal TB, Bohman P, et al. Metronidazole in the treatment of non-specific vaginitis (NSV). Br J Vener Dis 1984; 60: 171–174.
15. Morrison A, Polisena J, Husereau D, et al. The effect of English-language restriction on systematic review-based meta-analyses: A systematic review of empirical studies. Int J Technol Assess Health Care 2012; 28: 138–144.
16. Gregoire G, Derderian F, Le Lorier J. Selecting the language of the publications included in a meta-analysis: Is there a Tower of Babel bias? J Clin Epidemiol 1995; 48: 159–163.
17. Cook DJ, Guyatt GH, Ryan G, et al. Should unpublished data be included in meta-analyses? Current convictions and controversies. JAMA 1993; 269: 2749–2753.