{"paper_id":"3bd91878-0b82-4025-9bc3-525471167e50","body_text":"Osteoarthritis (OA) is the most prevalent form of musculoskeletal disease and\nis a major cause of impairment in an elderly population, with over 10% of the\nworld’s population over age 60 years reporting clinical problems associated\nwith OA( 1 ,  2 ) including severe pain and disability.( 3 ,  4 ) The prevalence of OA among\nwomen is greater than men, with rates of 14.9% versus 7.9% (ages 60-69), and 16.5\nversus 10.2 % (ages 70-79).( 5 ,  6 ) Sex hormones, particularly estrogen, are hypothesized\nto play an important role in the pathophysiology of OA in women( 7 ) through multiple pathways.( 8 – 11 )\nAdditionally, other reproductive factors such as parity and pregnancy may affect\nrisk for OA development due to factors such as weight gain, joint loading, and local\nand systemic inflammation.( 12 – 14 )\nThe relationship between reproductive factors and OA is complex and prior\nstudies have reported conflicting findings, which may be due to heterogeneity in\nsample size, methodology, populations, and demographics. Two large scale studies\nreported that parity and younger age at menarche were both positively associated\nwith either OA or proxies of OA.( 15 ).( 16 ) However, other studies have reported a\nnon-significant relationship between parity and radiographic OA( 17 ), as well as the age of menarche and menopause and the\nrisk of hip OA.( 18 ,  19 ) Additionally, a review of 16 studies found that there\nwas no association between female hormonal aspects and OA of the hand, hip, and\nknee.( 20 ) The literature on breastfeeding\nand OA is limited, but one study suggested that breastfeeding may be protective for\ncarpometacarpal joint (CMC) OA.( 21 )\nThe existing literature is conflicting, and most studies have included only a\nsubset of relevant reproductive factors. We aimed to study associations of\nreproductive factors with self-reported OA using the large, multi-ethnic\nWomen’s Health Initiative (WHI) in a retrospective cross-sectional format,\nusing the WHI’s rich dataset on reproductive history and exogenous estrogen\nuse.\n\nThe Women’s Health Initiative (WHI) baseline cohort includes\n161,808 post-menopausal women aged 50-79 years who were enrolled in 40 clinical\ncenters throughout the US; details have been published previously.( 22 ) Women were enrolled between 1993 and\n1998 into either an Observational Study (OS) cohort (93,676 women) or a Clinical\nTrial (CT) cohort (68,132 women). Women who self-reported rheumatoid arthritis,\nsystemic lupus erythematous, or ulcerative colitis were excluded from the study\ndue to possible overlap of symptoms, which may result in inaccurate self-report\nof OA. Of the 161,808 women enrolled in the WHI, 151,767 remained after\nexclusion for these conditions. We did not exclude women who underwent joint\nreplacement as this population likely included women with OA. The final cohort\nincluded 147,975 women who were not missing information on parity or type of\narthritis, and the parous cohort consisted of 130,331 women ( Figure 1 ).\nOsteoarthritis prevalence was self-reported on Form 30 of the WHI at\nbaseline. Participants were asked “Did your doctor ever say that you have\narthritis?” with the choices as “Yes” or\n“No.” For participants selecting “Yes,” they were\nthen asked “What kind of arthritis do you have?” with possible\nresponses of “rheumatoid arthritis,” “others/don’t\nknow,” and “missing.” Women who answered no to the first\nquestion were classified as not having osteoarthritis. Women were classified as\nhaving osteoarthritis if they answered yes to the first question and\n“Others/don’t know” to the second question. This method has\nbeen previously used in a WHI study on osteoarthritis, ethnicity, and BMI.( 23 )\nParity was assessed based on women’s response to the questions\n“Have you ever been pregnant?” and “How many live births\ndid you have?”. Women who answered that they had never been pregnant or\nthat they had no live births were classified as nulliparous. Women who reported\nat least one live birth were classified as parous. Age at menarche was\nself-reported through questionnaires. For women with no oophorectomy, age at\nmenopause was self-reported, defined by the age at which a woman last had any\nmenstrual bleeding or began using menopausal HT. For hysterectomy without\nbilateral oophorectomy, the age at menopause was self-reported based on the age\nat which a woman either began using HT or first had vasomotor symptoms. For\nwomen who had a hysterectomy without bilateral oophorectomy at age 50 years or\nolder but no use of HT or symptoms, the age at menopause was defined as the age\nat hysterectomy. If the algorithm defined an age at menopause as older than 60\nyears, it was recorded as 60 years.( 24 )\nBreastfeeding period was self-reported based on question “Thinking about\nall the children you breastfed, how many months total did you\nbreastfeed?”\nWe employed logistic regression to estimate the univariate association\nbetween osteoarthritis and reproductive factors that may influence estrogen\nexposure over a lifetime. The associations of the above reproductive factors\nwere further explored in a single multivariable model while adjusting for\npotential confounders. We investigated multicollinearity between the predictors\nprior to fitting the multivariable models. Of note, we were only able to study\nassociations rather than causality as this was a cohort study, and there was not\nan  a priori  experiment set up to investigate this relationship.\nWe adjusted for potential confounders that may affect the relationship between\nosteoarthritis and reproductive factors based on literature. Confounders\nincluded the following( 25 ): parity,\nbaseline HT use and duration in years, OCP use and duration in years (oral\ncontraceptive pills which may be combined estrogen + progestin, estrogen only,\nor progestin-only), age at WHI baseline, ethnicity (American Indian, black,\nHispanic, other, Pacific Islander, white), income (<$35,000,\n$35,000-99,000, >$100,000), insurance status (yes/no), current health\ncare provider (yes, no), smoking (never, past, current), alcohol use (never,\npast, current), body mass index (BMI defined as kg/m 2 ), history of\ndiabetes mellitus (yes, no), physical activity, and enrollment into OS versus\nCT. Physical activity was assessed through baseline questionnaires about\nintensity and duration of exercise, which was then converted into a measure of\nmetabolic equivalent (MET)-min/week (<100, 100-450, 450-1200,\n>=1200). Among parous women, we then analyzed odds ratios of\nosteoarthritis in relation to age at first birth, age at first pregnancy, number\nof pregnancies, number of live births, and breastfeeding duration in a\nmultivariable model adjusted for the potential confounders listed above. All\nconfounders were ascertained from baseline questionnaires, which were given at\nenrollment between 1993 and 1998.\nDue to missing self-reported information of some potential confounders\nat baseline in the WHI cohort, we used multiple imputation to allow us to\ninclude all women in the models. The number of women missing each variable is\ndisplayed in  Table 1 . For multiple\nimputation, we used the R package Multivariate Imputation by Chained Equations\n(MICE)( 26 ), which imputes each\nmissing value with a simulated plausible value until all missing values are\nimputed. We created 5 imputed datasets, ran our models on each dataset, and then\npooled the estimates from the imputed datasets. Additional details on the\nmissing data and multiple imputation approach are included in  Supplement 1 . We expect there to be\ndifferences from a multiple imputation-based approach and a complete-case\napproach, as validity for the latter relies on an assumption that the data are\nmissing completely at random, an unrealistic assumption for most studies in\nmedicine. The validity for the former (our approach) relies on an assumption\nthat the data are missing at random, a more practical and flexible assumption\nthat missingness is related to observed variables only. Analyses were conducted\nin R version 3.2.3. All statistical tests were two-sided and performed at the\nalpha = 0.05 level. P-values were not adjusted for multiple comparisons given\nsingle outcome and should be interpreted accordingly.\n\nThe baseline characteristics in our cohort are displayed in  Table 1A , stratified by parity status. In our cohort,\nthere were 130,331 (88.1%) parous women and 17,644 (11.9%) nulliparous women. Most\ncharacteristics were statistically significantly different between the two groups\ndue to large sample size, though many values were similar. The mean age in the\nparous cohort was 63.2 years versus 62.6 years in the nulliparous cohort. Both\ncohorts were primarily Caucasian. Reproductive characteristics for the parous cohort\nare displayed in  Table 1B .\nIn the entire cohort, 64,868 (43.8%) women reported osteoarthritis.  Table 2  displays the odds ratios of self-report\nof OA in relation to reproductive factors for the entire cohort. No significant\nclinical associations were found between measures of reproductive history and\nosteoarthritis due to small effect size. Younger age at menarche was associated with\nstatistically increased (but clinically insignificant) likelihood of OA (aOR 0.992,\n95% CI 0.984-0.999 for age 12; aOR 0.996, 95% CI 0.988-1.003 for age 13; and aOR\n0.998, 95% CI 0.990-1.007 for age >13 compared to age <12; global\np<0.001) in our adjusted analysis, while age at menopause was not associated\nwith OA (aOR 1.000 per year, 95% CI 0.997-1.002, p=0.851). History of parity was\nassociated with OA, aOR 1.017 (95% CI 1.009-1.026, p<0.001 compared to\nnon-parous). OA had clinically insignificant associations with both OCP use (aOR\n1.008 95% CI 1.001-1.016, p<0.01 compared to never users), and current use of\nHT [reference current users, 0.951 (95% CI 0.943-0.959) for never and aOR 0.981 (95%\nCI 0.972-0.989) for past users, global p<0.001]. Odds of OA did not show a\nclinically significant association with duration of OCPs (aOR 0.998 per year, 95% CI\n0.997-0.999 p<0.001) or HT (aOR 1.001 per year, 95% CI 1.000-1.002 p=0.012).\nWomen who had a history of hysterectomy had self-report OA aOR 1.013 (95% CI\n1.004-1.022) compared to no hysterectomy. In addition, women with unilateral\noophorectomy had risk of OA aOR 1.015 (95% CI 1.004-1.026, p<0.01) compared\nto no oophorectomy; however, women with history of bilateral oophorectomy were found\nto have decreased OA risk (aOR 0.987, 95% CI 0.978-0.995, p = 0.007 compared to\noophorectomy). Again, it is important to note all associations were clinically\ninsignificant.\nWe also separately investigated the relationship between osteoarthritis and\nadditional reproductive factors among parous women in  Table 3 . Age at first birth and pregnancy were not\nassociated with OA. The number of pregnancies and live births in relation was\nassociated with OA (global p<0.001 for both); however, no clear pattern was\nobserved with the number of pregnancies or births in relation to OA. Similarly, no\nclear pattern was observed with breastfeeding duration and OA.\n\nThe relationship between hormonal factors and OA is complex. Reproductive\nfactors are known to affect estradiol levels, and some prior studies have reported\nthat decreased estradiol levels are associated with increased risk of knee OA.( 27 ,  28 )\nEstradiol has been demonstrated to promote the health of skeletal muscle through\nmultiple pathways, including transforming growth factor-β and insulin-like\ngrowth factor-1 and 2.( 8 – 10 ) OA is more common in women than men and its\nincidence also rises sharply starting around the time of menopause, which also\nsuggests possible contribution of hormonal factors in its pathogenesis.( 5 ,  6 )\nHowever, despite these studies, the mechanisms underlying the relationship between\nestrogen (and other hormones) with OA remain heterogeneous and unclear and further\nunderstanding of clinical associations is warranted. Other reproductive factors such\nas parity and pregnancy may also affect OA development due to weight gain, joint\nloading, and local and systemic inflammation.( 12 – 14 )\nPrevious studies on the relationship between OA and reproductive factors\nhave reported conflicting findings, likely due to heterogeneity in sample size,\nmethodology, populations and demographics, and different methods of assessing OA\n(including self report, radiograph assessment, and proxies such as joint\nreplacement). Multiple studies have reported the relationship of OA with parity and\nage of menarche/menopause. A study of 4.6 million Danish patients revealed that\nparity was positively associated with OA hospitalizations, the risk of OA\nhospitalization (particularly for knee) increased with number of children, and all\nsubtypes of OA hospitalization were positively associated with time after the most\nrecent childbirth.( 15 ) The Million Women\nStudy prospectively analyzed the association between reproductive factors and the\nrisk of primary hip and knee replacement for OA among 1.3 million women in England\nand Scotland (1996-2001). Younger age at menarche (<12) was associated with\ngreater risk of hip and knee replacement for OA (OR1.09-1.15 depending on age\ncategory), while menopausal status and the age of menopause were not significantly\nassociated with hip or knee replacement for OA.( 16 ) In another study, increasing age of menarche was associated with\nreduced risk of total knee replacement due to primary OA.( 29 ) A separate study also reported that younger age at\nmenarche was associated with risk of radiographic hand OA.( 30 ) However, another study found that with regard to\nradiographic OA, parous women had higher odds of both joint space narrowing and\nosteophytes compared to nulliparous women, though neither reached statistical\nsignificance.( 17 ) Several other studies\nhave reported no significant association between the age of menarche or menopause\nand the risk of hip OA.( 18 ,  19 )\nThough the existing literature is conflicting as noted above, our analysis\nis consistent with prior large-scale studies in finding that younger age at menarche\nand history of parity are associated with statistically increased risk of OA,( 15 ,  16 ,\n 31 ,  32 ) though not all studies have reported these relationships.( 18 ,  21 ,\n 33 ) However, our findings were not\nclinically significant as compared to other studies that found effect sizes ranging\nfrom 5-20% as previously cited in this paragraph. Age at menarche and parity both\naffect lifetime exposure to estrogen; in addition, pregnancy and childbirth can also\ncause physiologic changes such as increased weight on joints which may impact OA\nrisk. However, in our study no clear trend existed for the number of pregnancies or\nlive births in relation to OA. We also found that a number of factors were not\nassociated with self-reported OA in the WHI, including age at menopause, first\nbirth, and first pregnancy; the literature findings on these associations have been\ninconsistent as above. Additionally, a systematic review of 16 studies found that\nthere was no association between female hormonal aspects and OA of the hand, hip,\nand knee.( 20 ) In our study, we did not find\nany clinically significant association with most other studies with odds ratios very\nclose to 1. The definition of clinical significance varies depending on the field\nand variables of interest, but is typically closer to the range of an effect size of\nat least 5-10%, which is larger than the findings in our study. Other articles cited\non this subject (see studies cited in  Discussion ) have typically reported effect sizes of greater than 5%.\nThis may be due to our large sample size as well as our ability to control for a\ncomprehensive set of confounders and reproductive factors, including hysterectomy\nand oophorectomy status.\nThe literature on hysterectomy and oophorectomy in relation to OA is very\nlimited and our study is one of the first to study this association, finding a\nstatistically but not clinically significant increased effect. Hysterectomy may\naffect OA due to complex hormonal effects; studies have shown that hysterectomy\nitself, even with ovarian conservation, may affect hormonal function and lead to\nearlier menopause and ovarian failure.( 34 )\nLimited studies have been mixed on the relationship between oophorectomy,\nhysterectomy, and OA ( 33 ,  35 ,  36 ); however,\nit is unclear if oophorectomy and HT were included as confounders in all studies.\nThese complex hormonal relationships of hysterectomy and oophorectomy in relation to\nOA warrant further investigation.\nOur study also found that use of OCPs and current use of HT were both found\nto be associated with increased OA self-report in WHI, though findings were not\nclinically significant. The literature on HT and OA is conflicting. The Million\nWomen Study of 1.3 million patients also reported that current use of postmenopausal\nHT was associated with significant increase in hip and knee replacement incidence,\nwhile OCPs were not associated with OA; mechanisms of these findings were\nunclear.( 16 ) A cross-sectional study of\n489 women also found that parity (but not HT or OC use) was independently associated\nwith greater patellar cartilage defects (OR 2.87).( 17 ) However, several studies have not found significant associations or\neven reductions of OA in relation to HT or OC use.( 37 ).( 36 ) ( 38 ) The Women’s Health Initiative has previously\nreported that women receiving estrogen-only therapy had significantly lower rates of\nany arthroplasty (as a proxy for severe OA), though associations were not\nsignificant for either hip or knee arthroplasty when examined separately. No\nassociation was found for estrogen-plus-progestin replacement and arthroplasty,\nsuggesting a possible effect of unopposed estrogen in relation to bone health.( 39 ) This is in contrast to our findings of HT\nbeing associated with increased self-report of OA. This may be due to the fact that\nwe studied any self-report of OA (as compared to OA that was clinically significant\nenough to result in arthroplasty). In addition, HT use may signify lower levels of\nestrogen due to hypoestrogenism symptoms and lower estradiol levels at baseline,\nwhich may reflect differences in the underlying populations rather than the effect\nof HT itself (as the timeline of when the OA developed in relation to the hormone\nuse is unknown in our cohort). Overall, the relationship between HT and OA is very\nconflicting in literature and warrants further study, including on durations and\ntypes of HT.\nIn regards to breastfeeding, we found no clinically significant associations\nwith odds ratios close to 1. Lactation is a low estrogen state, and the limited\nevidence is mixed on breastfeeding and osteoarthritis. A cross-sectional study of\n348 women reported that ever breastfeeding was found to be associated with a 63%\ndecrease in clinically verified carpometacarpal joints (CMC) OA. However, this\nassociation was not found to be consistent across sites or severity measures and may\nreflect false positive association.( 21 ) This\nis in contrast to an NHANES study which found that women who breastfed for one month\nor longer had a statistically significant 21% increased self-reported OA risk.( 40 ) Overall the literature on the effect of\nreproductive events on OA is inconsistent and of unclear clinical significance.\nThe strengths of our study include the large sample size and richness of the\ndataset for reproductive variables of interest. Additionally, we were able to adjust\nfor not only reproductive factors, but also known confounders of OA including BMI,\ndiabetes, as well as proxies of access of health care. We also examined OA directly\nrather than proxies of OA such as hospitalization or joint replacement. We were also\nable to include information on hysterectomy status and oophorectomy status, which\nhas rarely been studied in literature. Overall, very few studies have included such\na comprehensive set of reproductive and non-reproductive confounders in relation to\nOA in the same cohort.\nA major limitation of our study was that OA was self-reported, which is\nsubject to recall bias and inaccuracy. A prior WHI study reported that validation of\nself-reported data collection for OA has not been widely investigated, and another\nstudy found that a rheumatologist could confirm 81% of self-reported OA cases.( 23 ,  41 )\nWe attempted to adjust for this by excluding other conditions which may be confused\nwith OA, including rheumatoid arthritis, SLE, and Crohn’s Disease. The\nliterature definitions of OA vary widely and range from self-report to radiographic\nverification to joint replacement; compared to our study, objective measures such as\nradiographic OA or joint replacement may be more valid and objective, though may not\ncapture less severe OA cases or cases that were not radiographically verified. It is\nimportant to note that our findings can only be interpreted for self-report of any\nOA. In addition, our cohort was mostly Caucasian, limiting the generalizability of\nthe analysis. Another limitation was that the study was a retrospective\ncross-sectional format and therefore we do not have information on the timing of OA\ndevelopment in relation to different reproductive factors, which makes interpreting\nassociations more difficult. For reproductive factors occurring late in life (e.g.\nage at menopause), it is possible that OA developed prior to menopause and we cannot\nbe certain of the temporal relationship between the exposure and the outcome.\nHowever, the incidence of OA dramatically increases with age, so we would expect\nmost diagnoses to have occurred when women were older. Another limitation is that we\nonly had report of overall OA and not site-specific OA or information about OA\nseverity; it is important to note that OA risk factors likely differ by site. The\nfindings may be diluted as a result.\nIn conclusion, in the large, multi-ethnic Women’s Health Initiative,\nwe did not find reproductive factors to be clinically significant in relation to\nself-report of OA. The existing literature on reproductive factors in relation to OA\nis conflicting and heterogeneous; our study adds to the body of literature\nsuggesting that reproductive factors are likely not to have large clinical\nsignificance on the development of OA after adjustment for comprehensive risk\nfactors. Given the large economic and medical burden of OA in older women, this\nsubject warrants additional investigation in large prospective cohorts with control\nfor all relevant reproductive and hormonal factors. In particular, the relationships\nbetween hysterectomy, oophorectomy, and HT in relation to OA have been rarely\nstudied in literature. Additionally, other areas for future study include\nreproductive history in relation to biological mechanisms of OA, specific OA\nsubtypes, degrees of clinical severity of OA, ethnic differences in OA, and timing\nof OA development.","source_license":"CC-BY-4.0","license_restricted":false}