You are here

  • Home
  • Archive
  • Volume 73, Issue 11
  • Alcohol consumption as a preventive factor for developing rheumatoid arthritis: a dose-response meta-analysis of prospective studies

Article Text

Article menu
Clinical and epidemiological research
Extended report
Alcohol consumption as a preventive factor for developing rheumatoid arthritis: a dose-response meta-analysis of prospective studies
  1. Zhichao Jin1,
  2. Chun Xiang1,
  3. Qing Cai2,
  4. Xin Wei3,
  5. Jia He1
  1. 1Department of Health Statistics, Second Military Medical University, Shanghai, China
  2. 2Department of Rheumatology and Immunology, Changhai Hospital, Shanghai, China
  3. 3Renji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China
  1. Correspondence to Professor Jia He, Department of Health Statistics, Second Military Medical University, No. 800 Xiangyin Road, Shanghai 200433, China; hejia63{at}yahoo.com

Abstract

Objective To summarise the evidence regarding the dose-response association between alcohol consumption and risk of rheumatoid arthritis (RA).

Method Studies were identified from search of MEDLINE, Embase and Web of Science databases between 1 January 1946 and 10 April 2013, and from review of the conference abstracts and the reference lists of retrieved articles. Prospective studies that reported relative risks (RRs) with 95% CIs for the association between alcohol consumption and the risk of RA were included. Results from individual studies were pooled using a dose-response meta-analysis.

Results Up to 10 April 2013, 8 prospective studies contained 195 029 participants and 1878 RA cases were included. The results indicated that low to moderate alcohol consumption yielded a preventive effect on RA development (RR: 0.86; 95% CI 0.78 to 0.94), and provided some evidence of a non-linear relationship between alcohol consumption and risk of RA. Dose-response meta-analysis of the study data revealed that compared with that for no alcohol consumption, the adjusted RR was 0.93 (95% CI 0.88 to 0.98) for 3 g/day of alcohol consumption, 0.86 (95% CI 0.76 to 0.97) for 9 g/day, 0.88 (95% CI 0.78 to 0.99) for 12 g/day, 0.91 (95% CI 0.81 to 1.03) for 15 g/day, and 1.28 (95% CI 0.94 to 1.73) for 30 g/day. Subgroup analysis indicated that women who had low to moderate alcohol consumption had a 19% reduction in RA risk. Regardless of sex, a consistent low to moderate alcohol consumption for a period of at least 10 years was found to have a 17% reduction in RA risk.

Conclusions Low to moderate alcohol consumption inversely associated with the development of RA in a manner that appears to be dose-dependent, time-dependent and sex-dependent. Large prospective studies that investigate gene-environment interactions are required to further clarify the aetiology of RA.

  • Epidemiology
  • Rheumatoid Arthritis
  • Inflammation

https://doi.org/10.1136/annrheumdis-2013-203323

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Rheumatoid arthritis (RA), an autoimmune disease, characterised by chronic, destructive, debilitating arthritis that affects approximately 1% of the adult population,1 ,2 is hypothesised to result from the interaction between environmental and genetic factors.3 ,4 Regarding the possible environmental factors, those related to lifestyle are believed to play an important role.5 Alcohol consumption has been identified as a preventive factor, although the findings regarding its effect on RA development have been inconsistent. While the results of several observational studies suggest that moderate alcohol consumption may decrease the risk of RA,6–8 no clear-cut mechanism has been identified to explain this inverse relationship. One study that examined the association between alcohol consumption and markers of inflammation in preclinical patients with RA identified a U-shaped association between daily alcohol consumption and IL-6 levels prior to emergence of RA symptoms,9 suggesting the existence of a non-linear, possibly U-shaped or J-shaped, dose-response relationship between alcohol consumption and risk of RA.

    To further examine these findings and provide evidence of a hypothesised inverse association between low to moderate alcohol consumption and risk of RA, we performed a systematic review and meta-analysis to quantitatively assess the dose-response relationship between alcohol consumption and RA.

    Materials and methods

    Literature search

    A systematic literature search of MEDLINE, Embase and Web of Science databases for identification of articles published between 1 January 1946 and 10 April 2013 was performed by two investigators (CX and ZJ). The same search strategy was applied to Web of Science as that used for MEDLINE and Embase (see online supplementary search strategy) using the appropriate controlled vocabulary. No language restriction was imposed. The conference abstracts in the supplements of the ACR/ARHP Annual Scientific Meeting from 2006 to 2012 were also reviewed. The reference lists of the identified publications were reviewed for identification of additional studies. The headings and abstracts of the studies identified in the database were reviewed by two investigators (CX and ZJ) for identification of studies that met the following criteria: (1) use of prospective design, including cohort design or nested case-control design; (2) examination of alcohol consumption as the variable of interest; (3) determination of prevalence of RA, as identified by physicians and/or by use of the record linkage system, as the outcome of interest; and (4) reporting of the OR or relative risk (RR) of RA for high, moderate, or low alcohol consumption compared with no alcohol consumption, and their 95% CIs. All studies that used a retrospective case-control design were excluded based on recognition that (1) recall of alcohol consumption may be affected by recall bias after diagnosis or onset of disease, leading to denial of consumption or selective recall of presumed aetiological factors and (2) behavioural changes may occur after onset of symptoms that affect the reporting of consumption. However, nested case-control studies were not excluded, as these studies had collected dietary and other relevant data before diagnosis or onset of RA, and thus did not face the two limitations of retrospective case-control studies described above.

    Data extraction

    Data extraction was conducted by two investigators (ZJ and CX), and independently checked for accuracy by a third investigator (JH). For each included study, data regarding the title, author, publication year, country in which it was conducted, study design, sampling subjects, assessment of alcohol consumption, sample size, proportion of males, age of subjects, seropositive rate, duration of follow-up, covariates controlled for by matching or multivariable analysis, the number of cases/non-cases or person-year data and adjusted RR for each consumption category and its 95% CI were extracted. For studies that reported several multivariable-adjusted RRs, the effect estimate that was most fully adjusted for potential confounders was extracted. Study quality was assessed using the 9-star Newcastle-Ottawa Scale by two investigator (ZJ and QC).10 Online supplementary table S1 shows the definition of alcohol consumption used, and the means of conversion of categories within each study.

    Statistical analysis

    We examined the relationship between alcohol consumption and risk of RA on the basis of the adjusted RRs and 95% CIs reported in each study. Because the incidence of RA is low, the ORs in nested case-control studies approximate the RRs.11 ,12 Using the converted data in online supplementary table S1, meta-analyses were performed to compare the risk between high alcohol consumption and no alcohol consumption, and between low to moderate alcohol consumption and no alcohol consumption. A fixed-effects model was used to estimate the pooled RRs with 95% CIs if there was no evidence of heterogeneity; otherwise, a random-effect model was used.13 ,14 Dose-response meta-analysis was then conducted from the correlated natural log of RRs across categories of alcohol consumption.15 ,16 To derive the dose-response curve, alcohol consumption was modelled using restricted cubic splines with three knots at fixed percentiles (10, 50 and 90%) of the distribution. The p value for non-linearity was calculated by testing the null hypothesis that the coefficient of the second spline was equal to zero. The details of the methods used have been described by Larsson and Orsini.15 ,17 ,18

    The χ2 test and I2 statistic were used to explore the heterogeneity among studies. The Egger's regression test,19 Begg's test,20 and visual inspection of a funnel plot were performed to assess publication bias. Subgroup analyses by geographic area, sex, years of follow-up, type of drinking and serological status were performed to examine the relationship between alcohol consumption and RA risk. Sensitivity analyses were performed in two ways: first, by excluding those studies that met relatively fewer quality criteria of the Newcastle-Ottawa Scale (<7 stars); second, by excluding the studies that used a nested case-control design.

    Stata V.12.0 software (Stata, College Station, Texas, USA) was used for all analyses, and all statistical tests were two-sided; p<0.05 was considered an indication of statistical significance.

    Results

    Up to 10 April 2013, 2456 records were retrieved using the search strategy described. Review of the titles and abstracts according to the inclusion and exclusion criteria resulted in exclusion of 2422 articles. Reading of the full text of the remaining 34 articles for further evaluation resulted in the selection of eight studies, including five full-text articles,6 ,21–24 two abstracts,25 ,26 and one letter27 for subjection to meta-analysis. One identified study used the lowest consumption level (0.05–3.66 g/day) as the reference group.21 In an effort to avoid combining studies that were not comparable, we chose to exclude this study when comparing the risk of RA between high or low to moderate drinkers and non-drinkers. Online supplementary table S2 shows the studies that were excluded and the reason for their exclusion, and figure 1 shows the entire search process. Table 1 shows the general characteristics of the eight included studies which, together, had examined 195 029 participants and 1878 RA cases, and online supplementary table S3 presents the quality assessment of the included studies in detail.

    Figure 1
    Figure 1

    Selection of studies for inclusion in meta-analysis.

    View this table:
    Table 1

    Main characteristic of the included studies on alcohol consumption and risk of rheumatoid arthritis

    Effects of alcohol consumption on RA

    The summary RR of RA for a high level of alcohol consumption compared with no alcohol consumption (figure 2) was found not to be statistically significant (RR: 0.99; 95% CI 0.78 to 1.25). Low to moderate alcohol consumption was found to have a preventive effect for RA (RR: 0.86; 95% CI 0.78 to 0.94). No evidence of the existence of heterogeneity across studies was identified. As shown in figure 3, some evidence of a non-linear relationship between alcohol consumption and risk of RA was found (p=0.008). Compared with that for no alcohol consumption, the adjusted RR was 0.93 (95% CI 0.88 to 0.98) for 3 g/day of alcohol consumption, 0.86 (95% CI 0.76 to 0.97) for 9 g/day, 0.88 (95% CI 0.78 to 0.99) for 12 g/day, 0.91 (95% CI 0.81 to 1.03) for 15 g/day, and 1.28 (95% CI 0.94 to 1.73) for 30 g/day (see online supplementary table S4).

    Figure 2
    Figure 2

    Adjusted relative risks of rheumatoid arthritis for the high alcohol consumption versus no alcohol consumption and low to moderate alcohol consumption versus no alcohol consumption.

    Figure 3
    Figure 3

    Dose-response relationship between alcohol consumption and risk of rheumatoid arthritis.

    Subgroup analysis

    As shown in figure 2, compared with women who consume no alcohol, women who had low to moderate alcohol consumption were found to have a 19% reduction in RA risk (RR: 0.81, 95% CI 0.73 to 0.91). Regardless of sex, a consistent low to moderate alcohol consumption for a period of at least 10 years was found to have a 17% reduction in RA risk (RR: 0.83; 95% CI 0.75 to 0.92). As shown in online supplementary table S4, the subgroup analyses under the dose-response setting showed comparable results, which also found that generally, individuals consuming up to 15 g/day for a period of at least 10 years had a lower risk of RA than did non-drinkers.

    Sensitivity analysis

    As shown in online supplementary table S4, a dose-response trend similar to that identified by analysis of all eight studies emerged when three studies of low quality were excluded.22 ,25 ,27 The results of sensitivity analysis performed after excluding the studies that used a nested case-control design also indicated that low to moderate alcohol consumption is inversely correlated with risk of RA.

    Publication bias

    The funnel plot and Egger’s test (p=0.047) showed some evidence of publication bias in the comparison between low to moderate drinkers and non-drinkers (figure 4). No significant asymmetry of the funnel plot was detected in the comparison between heavy drinkers and non-drinkers.

    Figure 4
    Figure 4

    Funnel plot of log relative risk versus standard error of log relative risks.

    Discussion

    The findings of the meta-analysis described here indicate that alcohol consumption, but only at a low to moderate level, is inversely and significantly associated with the development of RA. Specifically, they indicate the existence of a J-shaped non-linear trend between alcohol consumption and risk of RA. This inverse association between alcohol consumption and risk of RA begins to lose its significance as consumption increases to more than 15 g/day, which is approximately 1.5 standard drinks in Europe and one standard drink in the USA.28

    A plausible explanation of a reduced risk of RA with low to moderate alcohol consumption is that alcohol may downregulate immune response and decrease the production of selected proinflammatory cytokines.6 ,29–32

    Compared with a newly published systematic review conducted by Scott et al,33 the results of the present study appear more reliable. Although Scott et al33 also found an inverse association between alcohol consumption and the presence of RA, they only got this result from retrospective case-control studies. The results of such studies cannot be used to identify cause–effect relationships, as inverse associations identified might simply reflect the fact that patients with RA drink less alcohol after disease onset, especially patients treated with methotrexate. By contrast, the results of prospective studies in which alcohol consumption was monitored in a cohort that was followed-up to determine which patients developed RA could be used to identify a causal relationship between alcohol consumption and risk of RA.34 Unlike the present study, Scott et al excluded abstracts and letters25–27 in their analysis. As the two abstracts included in the present study are from well-designed cohorts,35 ,36 excluding them may have resulted in insufficient statistical power and publication bias.

    This study had two strengths that likely increased the reliability and validity of the findings. First, only prospective studies were included, which could eliminate selection bias and recall bias that could be of concern in retrospective case-control studies. Second, the dose-response analysis included a wide range of alcohol consumption, which allows an accuracy assessment of the dose-response relationship between alcohol consumption and RA risk.

    This meta-analysis faced several potential limitations that must be considered when interpreting the results. First, a meta-analysis is not able to solve problems with confounding factors that could be inherent in the original studies. Although major potential confounders had been adjusted in most included studies, residual or unknown confounding cannot be excluded. Low to moderate drinkers may share, or be exposed to, a greater number of beneficial environmental factors compared with non-drinkers and heavy drinkers, such as a greater number of educational opportunities, greater ability to engage in physical activity, and access to a healthier diet. RA is believed to result from the interactions between environmental and genetic factors,37 but no data regarding genetic factors were contained in the primary aggregate results. Second limitation is misclassification of alcohol consumption, which was inevitable given that consumption of alcohol was self-reported in most studies and only two studies6 ,26 updated information on alcohol consumption during follow-up. Under-reporting of high alcohol consumption38 or changes in consumption during follow-up would most likely lead to underestimation of true association between alcohol consumption and risk of RA. Third, as in all meta-analyses of published studies, the possibility of publication bias was ever present. In the present study, some small studies with inverse association between alcohol consumption and risk of RA seemed to be suppressed. The presence of possible publication bias could have led to an underestimation of the beneficial effect of low to moderate alcohol consumption over RA. Fourth, results of several subgroups, for example, male subgroup, were based on a limited number of studies, and we, therefore, cannot rule out the possibility that insufficient statistical power may be present. Heterogeneity may be introduced because of methodological differences among studies. However, in this meta-analysis, no evidence of heterogeneity was found from the main results. The sensitivity analyses regarding methodological differences also have yielded consistent results with those from overall analysis.

    In summary, the results of this study suggested that low to moderate alcohol consumption may reduce the risk of developing RA, possibly in a dose-dependent, time-dependent and sex-dependent manner. These results have implications for public health measures, such as guidelines for lowering the risk of RA, which may be revised to recommend low to moderate alcohol consumption. However, any advice regarding alcohol consumption should be tailored to the individual risks and potential benefits. Large prospective studies that investigate the interactions between environmental factors and functional polymorphisms are now required to further clarify the aetiology of RA, and contribute to the ultimate aims of such research: establishment of an easily applicable prediction rule for the development of RA and provision of evidence-based recommendations to decrease the risk of RA.

    References

    1. ↵
      1. Gabriel SE,
      2. Crowson CS,
      3. O'Fallon WM
      . The epidemiology of rheumatoid arthritis in Rochester, Minnesota, 1955–1985. Arthritis Rheum 1999;42:415–20.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Alamanos Y,
      2. Drosos AA
      . Epidemiology of adult rheumatoid arthritis. Autoimmun Rev 2005;4:130–6.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Klareskog L,
      2. Stolt P,
      3. Lundberg K,
      4. et al
      . A new model for an etiology of rheumatoid arthritis: smoking may trigger HLA-DR (shared epitope)-restricted immune reactions to autoantigens modified by citrullination. Arthritis Rheum 2006;54:38–46.
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Kallberg H,
      2. Padyukov L,
      3. Plenge RM,
      4. et al
      . Gene-gene and gene-environment interactions involving HLA-DRB1, PTPN22, and smoking in two subsets of rheumatoid arthritis. Am J Hum Genet 2007;80:867–75.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Sugiyama D,
      2. Nishimura K,
      3. Tamaki K,
      4. et al
      . Impact of smoking as a risk factor for developing rheumatoid arthritis: a meta-analysis of observational studies. Ann Rheum Dis 2010;69:70–81.
      OpenUrlAbstract/FREE Full Text
    6. ↵
      1. Di Giuseppe D,
      2. Alfredsson L,
      3. Bottai M,
      4. et al
      . Long term alcohol intake and risk of rheumatoid arthritis in women: a population based cohort study. BMJ 2012;345:e4230.
      OpenUrlAbstract/FREE Full Text
    7. ↵
      1. Kallberg H,
      2. Jacobsen S,
      3. Bengtsson C,
      4. et al
      . Alcohol consumption is associated with decreased risk of rheumatoid arthritis: results from two Scandinavian case-control studies. Ann Rheum Dis 2009;68:222–7.
      OpenUrlAbstract/FREE Full Text
    8. ↵
      1. Maxwell JR,
      2. Gowers IR,
      3. Moore DJ,
      4. et al
      . Alcohol consumption is inversely associated with risk and severity of rheumatoid arthritis. Rheumatology (Oxford) 2010;49:2140–6.
      OpenUrlAbstract/FREE Full Text
    9. ↵
      1. Lu B,
      2. Solomon DH,
      3. Costenbader KH,
      4. et al
      . Alcohol consumption and markers of inflammation in women with preclinical rheumatoid arthritis. Arthritis Rheum 2010;62:3554–9.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Wells GA,
      2. Shea B,
      3. O'Connell D,
      4. et al
      . The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Ottawa, Canada: Department of Epidemiology and Community Medicine, University of Ottawa. http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp (accessed date Apr 20 2013).
    11. ↵
      1. Greenland S
      . Quantitative methods in the review of epidemiologic literature. Epidemiol Rev 1987;9:1–30.
      OpenUrlFREE Full Text
    12. ↵
      1. Zhang J,
      2. Yu KF
      . What's the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes. JAMA 1998;280:1690–1.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. DerSimonian R,
      2. Laird N
      . Meta-analysis in clinical trials. Control Clin Trials 1986;7:177–88.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Higgins JPT,
      2. Green S
      . Cochrane Handbook for Systematic Reviews of Interventions, Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. http://www.cochrane-handbook.org (accessed date Apr 20 2013).
    15. ↵
      1. Orsini N,
      2. Bellocco R,
      3. Greenland S
      et al. Generalized least squares for trend estimation of summarized dose-response data. Stata J 2006;6:40–57.
      OpenUrlWeb of Science
    16. ↵
      Greenland S, Longnecker MP. Methods for trend estimation from summarized dose-response data, with applications to meta-analysis. Am J Epidemiol 1992;135:1301–9.
    17. ↵
      1. Larsson SC,
      2. Orsini N,
      3. Wolk A
      . Dietary magnesium intake and risk of stroke: a meta-analysis of prospective studies. Am J Clin Nutr 2012;95:362–6.
      OpenUrlAbstract/FREE Full Text
    18. ↵
      1. Orsini N,
      2. Li R,
      3. Wolk A,
      4. et al
      . Meta-analysis for linear and nonlinear dose-response relations: examples, an evaluation of approximations, and software. Am J Epidemiol 2012;175:66–73.
      OpenUrlAbstract/FREE Full Text
    19. ↵
      1. Egger M,
      2. Davey Smith G,
      3. Schneider M,
      4. et al
      . Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629–34.
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Begg CB,
      2. Mazumdar M
      . Operating characteristics of a rank correlation test for publication bias. Biometrics 1994;50:1088–101.
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Bergstrom U,
      2. Jacobsson LTH,
      3. Nilsson JA,
      4. et al
      . Smoking, low formal level of education, alcohol consumption, and the risk of rheumatoid arthritis. Scand J Rheumatol 2013;42:123–30.
      OpenUrlCrossRefPubMed
    22. ↵
      1. Cerhan JR,
      2. Saag KG,
      3. Criswell LA,
      4. et al
      . Blood transfusion, alcohol use, and anthropometric risk factors for rheumatoid arthritis in older women. J Rheumatol 2002;29:246–54.
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Heliovaara M,
      2. Aho K,
      3. Knekt P,
      4. et al
      . Coffee consumption, rheumatoid factor, and the risk of rheumatoid arthritis. Ann Rheum Dis 2000;59:631–5.
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Rodriguez LA,
      2. Tolosa LB,
      3. Ruigomez A,
      4. et al
      . Rheumatoid arthritis in UK primary care: incidence and prior morbidity. Scand J Rheumatol 2009;38:173–7.
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Sundstrom B,
      2. Johansson I,
      3. Dahlqvist SR
      . Diet and other lifestyle related factors and the risk of developing rheumatoid arthritis [abstract]. Arthritis Rheum 2012;64(Suppl 10):2692.
      OpenUrl
    26. ↵
      1. Lu B,
      2. Daniel HS,
      3. Karen HC,
      4. et al
      . A Prospective study of alcohol consumption and risk of rheumatoid arthritis [abstract]. Arthritis Rheum 2011;63:2544.
      OpenUrlPubMed
    27. ↵
      1. Aho K,
      2. Heliovaara M
      . Alcohol, androgens and arthritis. Ann Rheum Dis 1993;52:897.
      OpenUrlFREE Full Text
    28. ↵
      ICAP Reports 5: What is a “Standard Drink”?. 1998. http://www.icap.org/LinkClick.aspx?fileticket=zo2TZaGGEoQ= (accessed date May 20 2013).
    29. ↵
      1. Mandrekar P,
      2. Catalano D,
      3. Dolganiuc A,
      4. et al
      . Inhibition of myeloid dendritic cell accessory cell function and induction of T cell anergy by alcohol correlates with decreased IL-12 production. J Immunol 2004;173:3398–407.
      OpenUrlAbstract/FREE Full Text
    30. ↵
      1. Waldschmidt TJ,
      2. Cook RT,
      3. Kovacs EJ
      . Alcohol and inflammation and immune responses: summary of the 2005 Alcohol and Immunology Research Interest Group (AIRIG) meeting. Alcohol 2006;38:121–5.
      OpenUrlCrossRefPubMedWeb of Science
    31. ↵
      1. Fan J,
      2. Edsen-Moore MR,
      3. Turner LE,
      4. et al
      . Mechanisms by which chronic ethanol feeding limits the ability of dendritic cells to stimulate T-cell proliferation. Alcohol Clin Exp Res 2011;35:47–59.
      OpenUrlCrossRefPubMed
    32. ↵
      1. Imhof A,
      2. Froehlich M,
      3. Brenner H,
      4. et al
      . Effect of alcohol consumption on systemic markers of inflammation. Lancet 2001;357:763–7.
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Scott IC,
      2. Tan R,
      3. Stahl D,
      4. et al
      . The protective effect of alcohol on developing rheumatoid arthritis: a systematic review and meta-analysis. Rheumatology (Oxford) 2013;52:856–67.
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. Frampton GK
      . Comment on: Alcohol consumption is inversely associated with risk and severity of rheumatoid arthritis. Rheumatology 2011;50:423–4.
      OpenUrlFREE Full Text
    35. ↵
      1. Norberg M,
      2. Wall S,
      3. Boman K,
      4. et al
      . The Vasterbotten Intervention Programme: background, design and implications. Glob Health Action 2010;3. doi: 10.3402/gha.v3i0.4643
    36. ↵
      1. Colditz GA,
      2. Hankinson SE
      . The Nurses’ Health Study: lifestyle and health among women. Nat Rev Cancer 2005;5:388–96.
      OpenUrlCrossRefPubMedWeb of Science
    37. ↵
      1. Padyukov L,
      2. Silva C,
      3. Stolt P,
      4. et al
      . A gene-environment interaction between smoking and shared epitope genes in HLA-DR provides a high risk of seropositive rheumatoid arthritis. Arthritis Rheum 2004;50:3085–92.
      OpenUrlCrossRefPubMedWeb of Science
    38. ↵
      1. Lemmens P,
      2. Tan ES,
      3. Knibbe RA
      . Measuring quantity and frequency of drinking in a general population survey: a comparison of five indices. J Stud Alcohol 1992;53:476–86.
      OpenUrlPubMedWeb of Science

    Supplementary materials

    Footnotes

    • Handling editor Tore K Kvien

    • ZJ, CX and QC contributed equally.

    • Contributors ZJ, JH and QC discussed and developed the question for this review. CX and ZJ carried out the searches. ZJ and CX assessed the eligibility of the studies for inclusion, extracted data and carried out all analysis. All authors were involved in interpreted and discussed results. ZJ wrote the first draft of this paper and it was reviewed by QC and JH. XW was involved in interpreted and discussed results in the first round of revision. All authors agreed on the final draft of this study. JH is the guarantor.

    • Funding National Natural Science Foundation (grant number 81001287), Leading Talents of Science in Shanghai 2010 (grant number 022), the key discipline construction of evidence-based public health in Shanghai (grant number 12GWZX0602).

    • Competing interests None.

    • Provenance and peer review Not commissioned; externally peer reviewed.