Erin S. LeBlanc, MD, MPH; Bernadette Zakher, MBBS; Monica Daeges, BA; Miranda Pappas, MA; Roger Chou, MD
This article was published online first at www.annals.org on 25 November 2014.
Disclaimer: The findings and conclusions in this document are those of the authors, who are responsible for its content, and do not necessarily represent the views of AHRQ. No statement in this report should be construed as an official position of AHRQ or the U.S. Department of Health and Human Services.
Acknowledgment: The authors thank Andrew Hamilton, MLS, MS, for conducting literature searches; Rongwei Fu, PhD, for statistical assistance; and Spencer Dandy, BS, for assistance with drafting this manuscript at the Oregon Health & Science University. The authors also thank Kevin Lutz, MFA, at the Center for Health Research for editorial assistance; AHRQ Medical Officers Robert McNellis, MPH, PA, Tina Fan, MD, MPH, and Tess Miller, DrPH; and U.S. Preventive Services Task Force Leads Linda Baumann, PhD, RN, Doug Owens, MD, MS, and Albert Siu, MD, MSPH.
Grant Support: By the Agency for Healthcare Research and Quality (contract HSSA 290-2007-10057-I).
Disclosures: Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M14-1659.
Requests for Single Reprints: Erin S. LeBlanc, MD, MPH, Center for Health Research, Kaiser Permanente, 3800 North Interstate Avenue, Portland, OR 97227; e-mail, firstname.lastname@example.org.
Current Author Addresses: Dr. LeBlanc: Center for Health Research, Kaiser Permanente, 3800 North Interstate Avenue, Portland, OR 97227.
Ms. Zakher, Ms. Daeges, Ms. Pappas, and Dr. Chou: Oregon Health and Science University, 3181 Southwest Sam Jackson Park Road, Mail code: BICC, Portland, OR 97239.
Author Contributions: Conception and design: E.S. LeBlanc, B. Zakher, M. Pappas, R. Chou.
Analysis and interpretation of the data: E.S. LeBlanc, B. Zakher, M. Pappas, R. Chou.
Drafting of the article: E.S. LeBlanc, M. Pappas, R. Chou.
Critical revision of the article for important intellectual content: E.S. LeBlanc, M. Pappas, R. Chou.
Final approval of the article: E.S. LeBlanc, M. Pappas, R. Chou.
Provision of study materials or patients: M. Daeges.
Statistical expertise: R. Chou.
Obtaining of funding: R. Chou.
Administrative, technical, or logistic support: M. Daeges, M. Pappas.
Collection and assembly of data: B. Zakher, M. Daeges, M. Pappas, R. Chou.
LeBlanc E., Zakher B., Daeges M., Pappas M., Chou R.; Screening for Vitamin D Deficiency: A Systematic Review for the U.S. Preventive Services Task Force. Ann Intern Med. 2015;162:109-122. doi: 10.7326/M14-1659
Download citation file:
Published: Ann Intern Med. 2015;162(2):109-122.
Vitamin D deficiency has been associated with adverse health outcomes.
To systematically review benefits and harms of vitamin D screening in asymptomatic adults.
Ovid MEDLINE (through the third week of August 2014), Cochrane Central Register of Controlled Trials, and Cochrane Database of Systematic Reviews.
Randomized trials of screening for and treatment of vitamin D deficiency and case–control studies nested within the Women's Health Initiative.
One investigator abstracted data, a second reviewed data for accuracy, and 2 investigators independently assessed study quality using predefined criteria.
No study examined the effects of vitamin D screening versus no screening on clinical outcomes. Vitamin D treatment was associated with decreased mortality versus placebo or no treatment (11 studies; risk ratio [RR], 0.83 [95% CI, 0.70 to 0.99]), although benefits were no longer seen after trials of institutionalized persons were excluded (8 studies; RR, 0.93 [CI, 0.73 to 1.18]). Vitamin D treatment was associated with possible decreased risk for having at least 1 fall (5 studies; RR, 0.84 [CI, 0.69 to 1.02]) and falls per person (5 studies; incidence rate ratio, 0.66 [CI, 0.50 to 0.88]) but not fractures (5 studies; RR, 0.98 [CI, 0.82 to 1.16]). Vitamin D treatment was not associated with a statistically significant increased risk for serious adverse events (RR, 1.17 [CI, 0.74 to 1.84]).
Variability across studies in 25-hydroxyvitamin D assays and baseline levels, treatment doses, use of calcium, and duration of follow-up.
Treatment of vitamin D deficiency in asymptomatic persons might reduce mortality risk in institutionalized elderly persons and risk for falls but not fractures.
Agency for Healthcare Research and Quality.
Vitamin D is obtained through food consumption and synthesis in the skin after ultraviolet (UV) B exposure (1). Researchers have reported associations between low 25-hydroxyvitamin D [25-(OH)D] levels and risk for fractures (2–6), falls (7, 8), cardiovascular disease (9–14), colorectal cancer (13–20), diabetes (13, 14, 21–29), depressed mood (13, 14, 30, 31), cognitive decline (13, 14), and death (13, 32).
Vitamin D deficiency is determined by measuring total serum 25-(OH)D concentrations (33). Measuring 25-(OH)D levels is complicated by the presence of multiple assays (34); evidence of intermethod and interlaboratory variability in measurement (35–43); and the lack of an internationally recognized, commutable vitamin D reference standard (44). Efforts to increase standardization are in progress (34, 44).
There is no consensus on optimal 25-(OH)D concentrations. Although experts generally agree that levels lower than 50 nmol/L (20 ng/mL) are associated with bone health (36, 45), disagreement exists about whether optimal 25-(OH)D levels are higher than this threshold (Table 1). According to NHANES (National Health and Nutrition Examination Survey) data from 2001 to 2006, 33% of the U.S. population was at risk for 25-(OH)D levels below 50 nmol/L (20 ng/mL) (47) and 77% had 25-(OH)D levels below 75 nmol/L (30 ng/mL) (48). Risk factors for low vitamin D levels include darker skin pigmentation (33), low vitamin D intake (49–51), little or no UVB exposure (49, 50, 52–54), and obesity (49–51, 55). Older age (49–53), female sex (49, 51, 52), low physical activity (49, 50, 53), low education attainment (48), and low health status (51, 54) were factors also associated with vitamin D deficiency in some studies.
Table 1. Summary of Current Opinions About Appropriate 25-(OH)D Level Cutoffs for Defining Vitamin D Deficiency and Associations Between These Cutoffs and Health Outcomes*
Vitamin D deficiency is treated by increasing dietary intake of food fortified with vitamin D or oral vitamin D treatment. Two commonly available vitamin D treatments (vitamin D3 [cholecalciferol] and vitamin D2 [ergocalciferol]) are available in several forms (for example, tablet and gel capsule), dosages (for example, 200 to 500 000 IU), and dosing regimens (for example, daily, weekly, monthly, or yearly) and can be given in combination with oral calcium (56, 57). Potential harms of vitamin D treatment include hypercalcemia, hyperphosphatemia, suppressed parathyroid hormone levels, and hypercalciuria (46, 58, 59). Although very high levels of vitamin D are associated with other potential harms, these events are rare with typical replacement doses (Table 1).
Screening for vitamin D deficiency can identify persons with low levels who might benefit from treatment. This report reviews the current evidence on vitamin D screening in asymptomatic adults to help the U.S. Preventive Services Task Force (USPSTF) develop a recommendation statement. Although the USPSTF has not previously issued recommendations on screening for vitamin D deficiency, it has made recommendations on vitamin D supplementation to prevent adverse health outcomes (for example, falls, fractures, cancer, and cardiovascular disease) in populations not necessarily vitamin D–deficient (that is, general populations who may or may not have been deficient) (60–63).
We developed a review protocol and analytic framework (Appendix Figure 1) that included the following key questions:
Numbers on figures indicate key questions. For a list of key questions, see the Methods section or Table 2.
1. Is there direct evidence that screening for vitamin D deficiency results in improved health outcomes?
1a. Are there differences in screening efficacy between patient subgroups?
2. What are the harms of screening (for example, risk for procedure, false positives, or false negatives)?
3. Does treatment of vitamin D deficiency using vitamin D lead to improved health outcomes?
3a. Are there differences in efficacy between patient subgroups?
4. What are the adverse effects of treatment of vitamin D deficiency using vitamin D?
4a. Are there differences in adverse effects between patient subgroups?
Detailed methods and data for this review are contained in the full report, including search strategies, inclusion criteria, abstraction and quality rating tables, and contextual questions (46). We developed our protocol using a standardized process after gathering input from experts and the public. The analytic framework focuses on direct evidence that screening for vitamin D deficiency improves important health outcomes (for example, death, falls, fractures, functional status, or risk for cancer) versus not screening. Further, the framework details evidence that treatment in persons found to have vitamin D deficiency is associated with improved health outcomes, harms resulting from screening or subsequent treatment, and how effects of screening and treatment vary in subgroups defined by demographic and other factors (for example, body mass index, UV exposure, and institutionalized status). We did not review the accuracy of vitamin D testing because of the lack of an accepted reference standard and studies reporting diagnostic accuracy.
For the purposes of this report, the term “vitamin D–deficient” refers to populations in which at least 90% of persons have 25-(OH)D levels of 75 nmol/L (30 ng/mL) or less. For studies that did not restrict enrollment to persons with 25-(OH)D levels of 75 nmol/L (30 ng/mL), we used the mean 25-(OH)D level plus the SD multiplied by 1.282 to approximate the 90th percentile to determine whether this level was at or below the 75-nmol/L (30-ng/mL) threshold. Because of uncertainty about what 25-(OH)D level constitutes deficiency, we stratified studies according to whether at least 90% of persons had levels less than 50 nmol/L (“<20 ng/mL” in this report) or at least 90% had levels less than 75 nmol/L (30 ng/mL) with at least 10% greater than 50 nmol/L (20 ng/mL) (“≤75 nmol/L [≤30 ng/mL]” in this report).
A research librarian searched Ovid MEDLINE (1946 through the third week of August 2014), Cochrane Central Register of Controlled Trials, and Cochrane Database of Systematic Reviews (through August 2014). We supplemented our electronic searches by reviewing reference lists of retrieved articles.
At least 2 reviewers independently evaluated each study to determine inclusion eligibility. For screening studies, we included randomized, controlled trials (RCTs) of screening for vitamin D deficiency versus no screening in healthy, asymptomatic adults (aged ≥18 years). For studies of the effectiveness of vitamin D treatment, we included RCTs of vitamin D treatment with or without calcium versus placebo or no treatment in vitamin D–deficient persons that reported health outcomes after at least 8 weeks of treatment. Because the Women's Health Initiative (WHI) is the largest RCT about vitamin D (64), we included data from nested case–control studies of WHI participants with known 25-(OH)D status.
We included English-language articles only and excluded studies published only as abstracts. We included studies conducted in the United States, Canada, United Kingdom, and other geographic settings generalizable to the United States. We excluded studies that specifically targeted populations with symptoms or conditions associated with vitamin D deficiency (for example, osteoporosis, history of nontraumatic fractures, or history of falls) or with medical conditions that increase a person's risk for deficiency (such as liver, kidney, or malabsorptive disease) because screening and treatment of vitamin D deficiency could be a component of medical management in these conditions. The summary of evidence search and selection is shown in Appendix Figure 2.
Summary of evidence search and selection.
25-(OH)D = 25-hydroxyvitamin D.
* Cochrane Central Register of Controlled Trials and the Cochrane Database of Systematic Reviews.
† Identified from reference lists or by hand-searching or suggested by experts.
‡ Studies that provided data and contributed to the body of evidence were considered included. Studies may have provided data for more than 1 key question or published article; 27 unique studies were included, and a total of 35 articles were included.
One investigator abstracted details about the study design, patient population, setting, screening method, interventions, analysis, follow-up, and results. A second investigator reviewed data for accuracy. Two investigators independently applied USPSTF criteria (65) to rate the quality of each study as good, fair, or poor. We resolved discrepancies through a consensus process. We excluded from data synthesis studies rated as poor quality. Those studies had 1 or more fatal flaws, including inadequate randomization or lack of intervention fidelity combined with postrandomization exclusions, high rates of withdrawals, and unclear randomization.
We assessed the aggregate internal validity (quality) of the body of evidence for each key question (good, fair, or poor) using methods developed by the USPSTF on the basis of the number, quality, and size of studies; consistency of results; and directness of evidence (65).
We conducted meta-analyses to calculate risk ratios (RRs) using the DerSimonian–Laird random-effects model (Review Manager, version 5.2; Cochrane Collaboration). Analyses were based on total follow-up (including time after discontinuation of vitamin D treatment). For falls per person, we calculated incidence rate ratios and assumed equal mean length of follow-up across treatment groups if these data were not reported. For analyses with between-study heterogeneity, we conducted sensitivity analyses using profile likelihood random-effects models (66). Rate ratio analysis and analyses using the profile likelihood model were done with Stata, version 12.0 (StataCorp). We performed sensitivity analyses restricted to RCTs, excluding the WHI subanalyses, and used odds ratios rather than RRs.
We assessed statistical heterogeneity using the chi-square test and I2 statistic (67). For all analyses, we stratified results by serum baseline 25-(OH)D level (<50 nmol/L [<20 ng/mL] vs. ≤75 nmol/L [≤30 ng/mL]). We performed additional analyses in which trials were stratified by institutionalized status, treatment regimen (vitamin D alone [vitamin D vs. placebo or no treatment, or vitamin D plus calcium vs. calcium alone] or vitamin D combined with calcium [vitamin D plus calcium vs. placebo or no treatment]), vitamin D dose (≤400 vs. >400 IU/d), duration of follow-up (≤12 vs. >12 months), and participant mean age (≤70 vs. >70 years).
This research was funded by the Agency for Healthcare Research and Quality (AHRQ) under a contract to support the work of the USPSTF. Investigators worked with USPSTF members and AHRQ staff to develop the scope, analytic framework, and key questions for this review. AHRQ had no role in study selection, quality assessment, or synthesis. AHRQ staff provided project oversight, reviewed the report to ensure that the analysis met methodological standards, and distributed the draft for peer review. The investigators are solely responsible for the content and the decision to submit it for publication.
No study evaluated clinical outcomes or harms in persons screened versus not screened for vitamin D deficiency.
Seven trials evaluated the effectiveness of vitamin D treatment (with or without calcium) in populations with at least 90% of persons with 25-(OH)D levels less than 50 nmol/L (20 ng/mL) (68–74). Nine trials and 1 nested case–control study evaluated effectiveness in populations with at least 90% of their population with levels of 75 nmol/L (30 ng/mL) or less (75–90) (Appendix Table). The mean age of the participants in these trials ranged from 37 to 85 years, and more than 70% of the studies enrolled only women. Mean body mass indices ranged from 24 to 36 kg/m2. The included studies were population-based or were conducted within outpatient clinics, academic institutions, and nursing or residential homes for elderly adults (considered institutionalized) in the United States or Europe. Ultraviolet exposure was not well-quantified in any study, and only 6 studies (64, 70, 71, 75, 82, 85) reported race. Of these, 1 study restricted enrollment to African Americans (70) and 83% to 100% of participants in the remaining 6 studies were white. Studies examined vitamin D3 at dosages ranging from 400 to 4800 IU/d or 8400 to 50 000 IU/wk. Five studies examined vitamin D3 treatment coadministered with calcium (1000 to 1200 mg/d), and 12 examined vitamin D3 treatment alone. Study duration ranged from 2 months to 7 years, and the assays these studies used to measure 25-(OH)D varied. Methodological shortcomings among these studies included unclear randomization and allocation concealment methods or blinding. Some studies had unclear intervention fidelity (that is, they did not record postintervention 25-[OH]D levels) or reported high attrition (>20%).
One good-quality trial, 9 fair-quality trials, and 1 fair-quality nested case–control study reported effects of vitamin D treatment (dose, 400 IU/d to 40 000 IU/wk) on mortality in vitamin D–deficient populations (n = 4126) (68–73, 77, 80, 82, 83, 89). Mortality was not a primary outcome in any study. No individual study reported a statistically significant reduction in mortality with vitamin D treatment versus placebo or no treatment, although the estimates were often imprecise because of very few events (68, 70–73, 77, 82). When data were pooled, vitamin D treatment with or without calcium was associated with decreased risk for mortality versus placebo or no treatment (RR, 0.83 [95% CI, 0.70 to 0.99]; I2 = 0%; absolute risk difference ranged from a reduction of 6 percentage points to an increase of 2 percentage points) (Appendix Figure 3).
Meta-analysis of effects of vitamin D treatment on mortality.
To convert ng/mL to nmol/L, divide by 0.40. 25-(OH)D = serum 25-hydroxyvitamin D.
* ≥90% of study participants had 25-(OH)D levels <20 ng/mL.
† ≥90% of study participants had 25-(OH)D levels ≤30 ng/mL, and ≥10% had 25-(OH)D levels ≥20 ng/mL.
‡ Included an institutionalized population.
§ This is a nested case–control study from the Women's Health Initiative calcium-vitamin D trial (64).
When analyses were stratified by institutionalized status, the risk reduction was limited to studies of older, institutionalized persons (3 studies; RR, 0.72 [CI, 0.56 to 0.94]; I2 = 0%; absolute risk reduction, 4 to 6 percentage points) (Figure 1) (69, 80, 83). The effect was not present in noninstitutionalized populations (8 studies; RR, 0.93 [CI, 0.73 to 1.18]; I2 = 0%) (68, 70–73, 77, 82, 89). In additional sensitivity analyses, the reduction in mortality occurred when studies exceeding 12 months whose populations had a mean age greater than 70 years were pooled. Stratification by baseline 25-(OH)D level, calcium use, or vitamin D dosage did not affect risk estimates.
Meta-analysis of effects of vitamin D treatment on mortality, by institutionalized status.
* This is a nested case–control study from the Women's Health Initiative calcium-vitamin D trial (64).
Four fair-quality trials and 1 nested case–control study examined the effects of 2 months to 7 years of vitamin D treatment (with or without calcium), 400 to 800 IU/d, on the risk for any type of fracture in vitamin D–deficient persons (n = 3551) (69, 74, 81, 84, 88). No individual study reported a statistically significant reduction in fracture risk with vitamin D treatment, including the largest study—a case–control analysis nested within the WHI calcium-vitamin D trial (88). The pooled estimate was close to 1 (5 trials; RR, 0.98 [CI, 0.82 to 1.16]; I2 = 32%) (Figure 2, top). Sensitivity analyses resulted in similar findings of no effect and did not decrease heterogeneity. Results were similar when only hip fracture risk was examined (4 trials; RR, 0.96 [CI, 0.72 to 1.29]; I2 = 46%) (Figure 2, bottom) (69, 74, 81, 88).
Meta-analysis of effects of vitamin D treatment on risk for any fracture (top) or hip fracture (bottom).
To convert ng/mL to nmol/L, divide by 0.40. 25-(OH)D = 25-hydroxyvitamin D.
† Population institutionalized.
‡ ≥90% of study participants had 25-(OH)D levels ≤30 ng/mL, with ≥10% with 25-(OH)D levels ≥20 ng/mL.
Five fair-quality trials examined the effects of 2 to 36 months of vitamin D treatment (with or without calcium), 800 IU/d, compared with control, on the risk for at least 1 fall (n = 1677) (69, 74, 76, 78, 84). Although the trials did not specifically recruit participants at high risk for frailty or those who had prior falls, these studies included persons who may have been at risk for falls based on older age (mean age >70 years) (69, 74, 76, 84), institutionalized status (69, 76), mobility problems (69, 76), or multiple comorbid conditions (69, 74, 76). In 2 studies, a proportion of patients had a history of falls (69, 76). Although the overall summary RR for experiencing at least 1 fall with vitamin D treatment was consistent with reduced risk (5 trials; pooled RR, 0.84 [CI, 0.69 to 1.02]) (Figure 3); the result was not statistically significant, and heterogeneity was high (I2 = 70%). Sensitivity analyses based on institutionalized status, baseline 25-(OH)D level, vitamin D dosage, study duration, and age did not reduce heterogeneity and resulted in similar estimates. Heterogeneity, however, was reduced to 0 when we excluded 2 trials of cotreatment with vitamin D and calcium (69, 78). Vitamin D treatment alone was associated with decreased risk for at least 1 fall (3 trials; RR, 0.65 [CI, 0.52 to 0.81]; I2 = 0%) (74, 76, 84).
Meta-analysis of effects of vitamin D treatment on risk for falls.
‡ ≥90% of study participants had 25-(OH)D levels ≤30 ng/mL, and ≥10% had 25-(OH)D levels ≥20 ng/mL.
§ The calculated risk ratio is different from the one reported by the study.
Five fair-quality trials examined the effect of vitamin D treatment (with or without calcium), 400 to 1000 IU/d, compared with control on the number of falls per person (n = 1399) (74, 76, 78, 84, 85). Vitamin D treatment was associated with a significant reduction in the number of falls per person versus placebo or no treatment (5 trials; incidence rate ratio, 0.66 [CI, 0.50 to 0.88]; I2 = 65%) (Figure 4). Although statistical heterogeneity was present, all estimates favored vitamin D treatment. Sensitivity analyses did not affect findings.
Meta-analysis of effects of vitamin D treatment on the number of falls per person.
To convert ng/mL to nmol/L, divide by 0.40. 25-(OH)D = 25-hydroxyvitamin D; PY = person-year.
One to 2 studies examined the effects of vitamin D (with or without calcium) on cancer risk (86, 90), type 2 diabetes mellitus risk (85, 87), psychosocial functioning and psychosocial disability (79, 91), and physical functioning (73). Findings either were mixed or showed no effect on these health outcomes.
None of the included trials were designed or powered to evaluate potential subgroup effects based on factors, such as sex, race, body mass index, or UV exposure. Data suggesting benefits of vitamin D treatment on mortality and falls seemed to be primarily limited to trials of older, often institutionalized, European women (69, 80, 83).
Twenty-four trials evaluated harms associated with vitamin D treatment (with or without calcium) in vitamin D–deficient populations aged 31 to 85 years (n = 4722) (Appendix Table) (68–73, 75–77, 79, 80, 82, 83, 85, 92–103). Vitamin D treatment (mostly D3 formulation) was given at doses of 400 to 7000 IU/d or 8400 to 54 000 IU/wk for 6 weeks to 4 years. Nineteen trials evaluated the vitamin D treatment alone, and 5 evaluated vitamin D with calcium. Methodological shortcomings included unclear randomization procedure; inadequate or unclear masking of assessors, providers, or participants; high attrition; and no clear statement that adverse events were a prespecified outcome.
Appendix Table. Studies of Effectiveness and Harms of Vitamin D Treatment
We found no difference between treatment with vitamin D and placebo or no treatment in risk for any adverse event (n = 1332; 7 trials), serious adverse events (n = 1401; 7 trials; RR, 1.17 [CI, 0.74 to 1.84]), withdrawals due to adverse events (n = 938; 5 trials; RR, 0.90 [CI, 0.36 to 2.24]), hypercalcemia (n = 3172; 16 studies; RR, 1.05 [CI, 0.57 to 1.94]), kidney stones (n = 1608; 7 trials, with no kidney stones reported in any trial), or gastrointestinal symptoms (n = 1201; 4 trials; RR, 0.84 [CI, 0.44 to 1.58]). The studies were not designed to evaluate whether harms differ according to demographic or other clinical characteristics.
The evidence reviewed in this report is summarized in Table 2. We found no direct evidence on effects of screening for vitamin D deficiency versus no screening on clinical outcomes. In persons with low vitamin D levels, vitamin D treatment was associated with decreased risk for death, but effects were no longer present when 3 trials of older institutionalized women were excluded from the analysis (69, 80, 83). Vitamin D treatment was associated with a nonsignificant reduction in the risk for 1 or more falls and a significantly reduced overall burden of falls, which is measured by the number of falls per person. This potential discrepancy seems largely attributable to 1 trial that was conducted in an institutionalized population with a high comorbidity burden; the trial reported a rate ratio for falls per person as its primary outcome that was lower than the risk for at least 1 fall (0.46 [CI, 0.28 to 0.76] and 0.75 [CI, 0.41 to 1.37], respectively) (76). The risk estimates were similar in 3 other trials that reported both risk for falls and the rate of falls per person (74, 78, 84). Data were limited (≤2 studies) on the effect of vitamin D on other outcomes, such as cancer risk, type 2 diabetes mellitus risk, psychosocial functioning, disability, and physical functioning. Vitamin D treatment did not seem to be associated with increased risk for harms, although few trials were designed to specifically address harms and harms reporting was often suboptimal. Evidence to evaluate subgroup effects on the basis of such factors as race, sex, age, or risk factors for vitamin D deficiency was very limited and precludes us from drawing reliable conclusions.
Table 2. Summary of Evidence for Screening for Vitamin D Deficiency in Asymptomatic Adults
An important limitation of the evidence is that no study specifically evaluated the effect of treatment of screen-detected vitamin D deficiency, which potentially limits their applicability to screening settings. Although we excluded studies that selected patients with conditions and outcomes associated with vitamin D deficiency, symptoms were not reported, which makes it difficult to know whether patients were truly asymptomatic. In addition, baseline 25-(OH)D levels, dosages used, use of calcium cosupplementation, and duration of follow-up varied among these studies. Sensitivity and stratified analyses on these factors, however, did not affect conclusions.
The included studies also used various vitamin D assays, and we cannot precisely determine how assay variability affected findings given the lack of a reference standard to estimate diagnostic accuracy. In general, differential classification due to assay variability is likely to affect persons with levels close to the threshold used to define vitamin D deficiency. In studies of vitamin D treatment, misclassification would attenuate estimates of treatment benefit because some persons who are not vitamin D–deficient would be classified and treated as such. These patients would also be subjected to unnecessary treatment and associated harms.
For this review, we required that participants in treatment studies be vitamin D–deficient. Previous USPSTF reviews on vitamin D evaluated supplementation in persons who were or were not deficient and could be at risk for a particular condition or outcome (104–106). On the basis of these reviews, the USPSTF made recommendations about vitamin D supplementation in persons whose deficiency status is unknown or are at risk for particular conditions. The USPSTF recommended vitamin D supplementation for community-dwelling adults aged 65 years or older at increased risk for falls regardless of 25-(OH)D status (60). The USPSTF recommended against low-dose supplementation with vitamin D (≤400 IU) and calcium (≤1000 mg) to reduce fracture risk in noninstitutionalized populations and concluded that data on the effects of higher doses were insufficient (62). The USPSTF also concluded that data were insufficient about the effects of vitamin D supplementation on cardiovascular disease and cancer risk (63). Previous reviews for the USPSTF found that harms were generally low (104–106). Prior systematic reviews noted that the WHI calcium-vitamin D trial found a significantly increased risk for kidney stones (64). We did not include these results because the risk for stones was not reported for women with low 25-(OH)D levels.
Our review had limitations. We excluded non–English-language articles and studies published only as abstracts, and we could not formally assess for publication bias because of the small number of studies. Some pooled analyses were based on small numbers of studies or were characterized by the presence of statistical heterogeneity. In these cases, the DerSimonian–Laird random-effects model may result in CIs that are too narrow (107). Therefore, we performed sensitivity analyses using the profile likelihood method that resulted in similar findings. We also focused on the effects of vitamin D treatment in patients similar to those who would be identified through a screening program. As such, we excluded studies that targeted populations for which vitamin D might be considered a treatment option or with particular medical conditions associated with vitamin D deficiency, even if the participants had low 25-(OH)D levels. On the basis of these criteria, we excluded trials that required participants to have osteoporosis or osteopenia (4 studies [108–111]), risk factors for falls (5 studies [112–116]), prediabetes (1 study ), heart failure (2 studies [118, 119]), or tuberculosis (1 study ). In those trials, vitamin D treatment did not reduce fracture risk in those with a history of fractures. Treatment reduced risk for falls in persons who had a history of falls (112) but not in those with a recent hip fracture (111) or at least 1 health problem or functional limitation (114).
A trial of screening for vitamin D in a diverse population would be the ideal way to evaluate benefits and harms. Greater standardization in vitamin D assays is needed for this study to be most informative. In addition, given the lack of consensus about what level of 25-(OH)D (for example, <50 vs. <75 nmol/L [<20 vs. <30 ng/mL]) defines deficiency (36, 45, 121–124), future studies of treatment should stratify results according to the baseline vitamin D level. Definitions of vitamin D deficiency may need to take into account potential racial differences in total 25-(OH)D levels relative to bioavailable levels (99).
In conclusion, no study directly examined the benefits and harms of screening for vitamin D deficiency. Based on limited evidence in persons not known to have conditions associated with vitamin D deficiency, treating this deficiency with vitamin D may be associated with decreased risk for death in institutionalized elderly adults and a reduction in the average number of falls but not fractures. Future research is needed to reduce assay variability; determine appropriate thresholds for vitamin D deficiency; and clarify effects of screening, subsequent treatment, and the subpopulations most likely to benefit.
The In the Clinic® slide sets are owned and copyrighted by the American College of Physicians (ACP). All text, graphics, trademarks, and other intellectual property incorporated into the slide sets remain the sole and exclusive property of the ACP. The slide sets may be used only by the person who downloads or purchases them and only for the purpose of presenting them during not-for-profit educational activities. Users may incorporate the entire slide set or selected individual slides into their own teaching presentations but may not alter the content of the slides in any way or remove the ACP copyright notice. Users may make print copies for use as hand-outs for the audience the user is personally addressing but may not otherwise reproduce or distribute the slides by any means or media, including but not limited to sending them as e-mail attachments, posting them on Internet or Intranet sites, publishing them in meeting proceedings, or making them available for sale or distribution in any unauthorized form, without the express written permission of the ACP. Unauthorized use of the In the Clinic slide sets will constitute copyright infringement.
Author Insight Video - Erin S. LeBlanc, MD, MPH
Dr Kirti Kain
Senior Clinical Lecturer, University of Leeds
December 10, 2014
Total versus bioavailable active Vitamin D
The USPSTF concludes that the current evidence is insufficient to assess the balance of benefits and harms of screening for vitamin D deficiency in asymptomatic adults . Current evidence is from studies whereby supplementation (variable doses of either vitamin-D2 or D3 with or without calcium) has been monitored mainly by the measurement of total concentrations of 25OHD. Future research studies should focus on assessing bioavailable vitamin D instead of total concentrations of 25OHD . It is known that concentrations of total status 25OHD as well as 1,25-dihydroxyvitamin-D measured routinely are different from bioavailable vitamin-D . Free and bioavailable vitamin-D is dependent on the vitamin-D binding protein and ethnicity  . In addition we need evidence of techniques of increasing bioavailable vitamin-D. One such technique could be increased outdoor physical activity which might possibly increase biosynthesis and bioavailable vitamin D3 circumventing hypervitaminosis D  .It is possible that harmful effects of hypervitaminosis D are solely due to excess biosynthesized sequestrated vitamin D as a result of inappropriate oral supplementations and not being converted to active bioavailable vitamin D . Excess Vitamin D is arteriotoxic and it causes elastocalcinosis which induces destruction of elastic fibers, which leads to arterial stiffness  and induces arterial calcification through up regulation of 1,25(OH)2D3 receptor and increased calcium uptake in smooth muscle cells of the arteries [7;8]. Research resources are finite in these times of austerity hence they ought to be allocated appropriately. Robust and pertinent evidence is needed to formulate educational and interventional policies that can be implemented to prevent the global public health problem of decreased bioavailable vitamin-D associated cardio-metabolic diseases, autoimmune and neoplastic conditions. References 1 LeFevre ML: Screening for Vitamin D Deficiency in Adults: U.S. Preventive Services Task Force Recommendation Statement. Ann Intern Med 2014. 2 Wong MS, Leisegang MS, Kruse C, Vogel J, Schurmann C, Dehne N, Weigert A, Herrmann E, Brune B, Shah AM, Steinhilber D, Offermanns S, Carmeliet G, Badenhoop K, Schroder K, Brandes RP: Vitamin D Promotes Vascular Regeneration. Circulation 2014. 3 Holick MF: Bioavailability of vitamin D and its metabolites in black and white adults. N Engl J Med 2013;369:2047-2048. 4 Kain K: Circulation. In press. 5 Wortsman J, Matsuoka LY, Chen TC, Lu Z, Holick MF: Decreased bioavailability of vitamin D in obesity. Am J Clin Nutr 2000;72:690-693. 6 Jegger D, da SR, Jeanrenaud X, Nasratullah M, Tevaearai H, von Segesser LK, Segers P, Gaillard V, Atkinson J, Lartaud I, Stergiopulo N: Ventricular-arterial coupling in a rat model of reduced arterial compliance provoked by hypervitaminosis D and nicotine. Am J Physiol Heart Circ Physiol 2006;291:H1942-H1951. 7 Rajasree S, Umashankar PR, Lal AV, Sarma PS, Kartha CC: 1,25-dihydroxyvitamin D3 receptor is upregulated in aortic smooth muscle cells during hypervitaminosis D. Life Sci 2002;70:1777-1788. 8 Rajasree S, Rajpal K, Kartha CC, Sarma PS, Kutty VR, Iyer CS, Girija G: Serum 25-hydroxyvitamin D3 levels are elevated in South Indian patients with ischemic heart disease. Eur J Epidemiol 2001;17:567-571.
Mark J Bolland, Andrew Grey, Ian R Reid
University of Auckland
February 4, 2015
Missing studies from vitamin D meta-analyses
The number of meta-analyses of vitamin D supplements for falls and fractures is about twice the number of randomized trials that have been carried out. Differences in conclusions of these meta-analyses are largely due to the methods adopted, including the choice of studies included (1, 2). In their review for the US Preventative Services Task Force (USPSTF), LeBlanc and colleagues assessed the effectiveness of vitamin D supplementation on mortality, falls, and fractures in vitamin D deficiency, including 11, 5 and 5 trials for each outcome respectively (3). In contrast, we included 38, 20 and 23 trials respectively, in meta-analyses of vitamin D supplementation for these conditions (4, 5). The differences in study inclusion are largely due to the requirement of LeBlanc and colleagues that baseline 25 hydroxyvitamin D (25OHD) was measured in all participants. Their aim was to include only studies in which 90% of participants had 25OHD <75 nmol/L, but random sampling of baseline 25OHD is sufficient to assess this criterion. Baseline 25OHD, either in a sample or in all participants, was reported in the majority (34/42) trials in our meta-analyses, with 25/32 (78%) reporting mean baseline 25OHD <50 nmol/L, meaning it is very likely that 90% of participants had 25OHD <75 nmol/L.Inconsistency in study and participant inclusion is an additional consequence of the methodology used by LeBlanc and colleagues. Of two studies carried out by the same investigators in the same population group (6, 7), one was included with 25OHD measured in all participants (7), whereas the other was excluded because 25OHD was only measured in a subset of participants (6), even though mean baseline 25OHD was similar in the two studies. Likewise, a small subset of participants in two studies (8, 9) were included in this meta-analysis because they were selected to have baseline 25OHD measured whereas the majority of participants in both studies were excluded.The upshot is that the meta-analyses by LeBlanc and colleagues contain few events and participants, and do not include the majority of studies with fracture or falls as the primary endpoint. The same issue of non-inclusion of eligible studies (2, 10) has occurred in previous reviews on vitamin D for the USPSTF (11, 12). Consequently, conclusions based on these reviews may not be reliable.References1. Bolland MJ, Grey A, Reid IR. Differences in overlapping meta-analyses of vitamin d supplements and falls. J Clin Endocrinol Metab. 2014;99(11):4265-72.2. Bolland MJ, Grey A. A case study of discordant overlapping meta-analyses: vitamin d supplements and fracture. PLoS One. 2014;9(12):e115934.3. LeBlanc ES, Zakher B, Daeges M, Pappas M, Chou R. Screening for vitamin d deficiency: a systematic review for the u.s. Preventive services task force. Ann Intern Med. 2015;162(2):109-22.4. Bolland MJ, Grey A, Gamble GD, Reid IR. The effect of vitamin D supplementation on skeletal, vascular, or cancer outcomes: a trial sequential meta-analysis. Lancet Diabetes Endocrinol. 2014;2(4):307-20.5. Bolland MJ, Grey A, Gamble GD, Reid IR. Vitamin D supplementation and falls: a trial sequential meta-analysis. Lancet Diabetes Endocrinol. 2014;2(7):573-80.6. Chapuy MC, Arlot ME, Duboeuf F, Brun J, Crouzet B, Arnaud S, et al. Vitamin D3 and calcium to prevent hip fractures in the elderly women. N Engl J Med. 1992;327(23):1637-42.7. Chapuy MC, Pamphile R, Paris E, Kempf C, Schlichting M, Arnaud S, et al. Combined calcium and vitamin D3 supplementation in elderly women: confirmation of reversal of secondary hyperparathyroidism and hip fracture risk: the Decalyos II study. Osteoporos Int. 2002;13(3):257-64.8. Lips P, Graafmans WC, Ooms ME, Bezemer PD, Bouter LM. Vitamin D supplementation and fracture incidence in elderly persons. A randomized, placebo-controlled clinical trial. Ann Intern Med. 1996;124(4):400-6.9. Karkkainen MK, Tuppurainen M, Salovaara K, Sandini L, Rikkonen T, Sirola J, et al. Does daily vitamin D 800 IU and calcium 1000 mg supplementation decrease the risk of falling in ambulatory women aged 65-71 years? A 3-year randomized population-based trial (OSTPRE-FPS). Maturitas. 2010;65(4):359-65.10. Bolland MJ, Grey A, Reid IR. Vitamin and mineral supplements in the primary prevention of cardiovascular disease and cancer. Ann Intern Med. 2014;160(9):655-6.11. Chung M, Lee J, Terasawa T, Lau J, Trikalinos TA. Vitamin D With or Without Calcium Supplementation for Prevention of Cancer and Fractures: An Updated Meta-analysis for the U.S. Preventive Services Task Force. Ann Intern Med. 2011;155(12):827-38.12. Fortmann SP, Burda BU, Senger CA, Lin JS, Whitlock EP. Vitamin and mineral supplements in the primary prevention of cardiovascular disease and cancer: An updated systematic evidence review for the U.S. Preventive Services Task Force. Ann Intern Med. 2013;159(12):824-34.
Toshihiro Sugiyama, Yoon Taek Kim, Hiromi Oda
Saitama Medical University
February 16, 2015
Risk reduction of falls but not fractures by treatment of vitamin D deficiency
A systematic review for the U.S. Preventive Services Task Force by LeBlanc and colleagues (1) concludes that treatment of vitamin D deficiency in asymptomatic adults might reduce risk for falls but not fractures. This requires reasonable explanation, because a fall is one of the strongest risk factors for a fracture. Here we would like to present an evidence-based mechanistic insight.Muscle force is related to bone strength and the risk reduction of falls but not fractures (1) theoretically means that treatment of vitamin D deficiency reduces fall risk by improving balance (rather than muscle force) but does not change fracture risk by impairing bone strength. The former would be consistent with clinical evidence, while the latter could be associated with skeletal adaptation to mechanical environment (2-5). There is a yield force at which a bone begins to deform plastically and normal physical activity causes the pre-yield “elastic” deformation (strain) of bone. A decrease in bone quality associated with mineral induces an increase in the “elastic” deformation, while the skeleton responds to mechanical environment to maintain the resultant strain of bone. Consequently, mineral-related impairment of bone quality can be compensated by mechanical strain-related feedback control and might act to improve bone fragility if compensated efficiently (3). For example, patients with hypophosphatemic rickets/osteomalacia have lower quality and higher quantity of bone (2). Vitamin D deficiency impairs bone quality and children with nutritional rickets would also have bigger long bones (4). Furthermore, a recent study in children with cerebral palsy showed an inverse correlation between serum levels of 25-hydroxyvitamin D [25-(OH)D] and Z-scores for areal bone mineral density (BMD) in the distal femur (4). Notably, the latest meta-analysis in adults found that the effects of daily supplementation with 800 IU or more of vitamin D on areal BMD were lower than that with less than 800 IU of vitamin D in the lumbar spine and potentially in the femoral neck, but not in the forearm (5).Finally, a number of observational studies have shown an association between lower levels of 25-(OH)D and higher incidences of fracture in adults. The present conclusion (1), however, suggests that confounding biases the association. Vitamin D status is strongly influenced by sunlight exposure associated with outdoor activity while mechanical loading from habitual physical activity is the primary determinant of bone strength, implying that higher incidences of fracture could result from lower levels of physical activity rather than 25-(OH)D.References1. LeBlanc ES, Zakher B, Daeges M, Pappas M, Chou R. Screening for vitamin D deficiency: a systematic review for the U.S. Preventive Services Task Force. Ann Intern Med. 2015;162:109-22. 2. Sugiyama T, Kim YT, Oda H. Osteoporosis therapy: a novel insight from natural homeostatic system in the skeleton. Osteoporos Int. 2015;26:443-7.3. Sugiyama T, Torio T, Sato T, Matsumoto M, Kim YT, Oda H. Improvement of skeletal fragility by teriparatide in adult osteoporosis patients: a novel mechanostat-based hypothesis for bone quality. Front Endocrinol. 2015;6:6.4. Sugiyama T, Yoshioka H, Sakaguchi K, Kim YT, Oda H. An evidence-based perspective on vitamin D and the growing skeleton. Osteoporos Int. 2015 doi:10.1007/s00198-014-2975-z5. Sugiyama T, Tanaka S, Miyajima T, Kim YT, Oda H. Vitamin D supplementation and fracture risk in adults: a new insight. Osteoporos Int. 2014;25:2497-8.
Erin S. LeBlanc, MD, MPH (1,2), Roger Chou, MD (2,3), Miranda Pappas, MA (2)
1. Center for Health Research Kaiser Permanente Northwest Portland, OR 2. Pacific Northwest Evidence-based Practice Center, Oregon Health & Science University, Portland, OR 3. Department o
March 14, 2015
The purpose of our systematic review was to determine if screening for vitamin D deficiency in asymptomatic persons improved health outcomes. Therefore, we determined a priori that we would only include studies of populations who had documented vitamin D deficiency and were not selected based on a history of osteoprorosis, prior fractures, or falls. Thus, we included fewer studies than the meta-analyses by Bolland et al., (1, 2) which also included trials of participants with normal vitamin D levels and with osteoporosis and prior fractures or falls. We did not exclude any study solely because only a random sample of the population had vitamin D deficiency according to our definition (90% of people with levels less than 30 ng/mL). Either the subsample did not meet this deficiency definition, or there was another reason that the study was excluded (e.g., patients with prior fractures or falls). In the specific example mentioned by Bolland, Grey, and Reid (3, 4), the subsample of participants did not meet our criteria for deficiency although when all participants were measured, over 90% were deficient. Although some of our analyses had relatively few events, expanding inclusion to clinically heterogeneous populations that are not of interest in order to increase statistical power would not have been appropriate.We agree with Kain that research on the role of bioavailable vitamin D levels is important to better understand the effects of vitamin D treatment on clinical outcomes. We also agree with Sugiyama, Kim, and Oda that research is needed on mechanisms for how vitamin D might prevent falls but not fracture.In response to an editorial by Heaney and Armas (5), we explicitly defined the scope of the review prior to starting the work. We addressed the factors brought up in the editorial as potentially impacting estimates in sensitivity and stratified analyses. Further stratifying or restricting the analysis, as suggested by Heaney and Armas, would only result in even less evidence to support benefits of vitamin D treatment.Reference List1. Bolland MJ, Grey A, Gamble GD, Reid IR. Vitamin D supplementation and falls: a trial sequential meta-analysis. Lancet Diabetes Endocrinol 2014 Jul; 2(7):573-580.2. Bolland MJ, Grey A, Gamble GD, Reid IR. The effect of vitamin D supplementation on skeletal, vascular, or cancer outcomes: a trial sequential meta-analysis. Lancet Diabetes Endocrinol 2014 Apr; 2(4):307-320.3. Chapuy MC, Arlot ME, Duboeuf F, Brun J, Crouzet B, Arnaud S, Delmas PD, Meunier PJ. Vitamin D3 and calcium to prevent hip fractures in the elderly women. N Engl J Med 1992 Dec 3; 327(23):1637-1642.4. Chapuy MC, Pamphile R, Paris E, Kempf C, Schlichting M, Arnaud S, Garnero P, Meunier PJ. Combined calcium and vitamin D3 supplementation in elderly women: confirmation of reversal of secondary hyperparathyroidism and hip fracture risk: the Decalyos II study. Osteoporos Int 2002 Mar; 13(3):257-264.5. Heaney RP, Armas LA. Screening for vitamin d deficiency: is the goal disease prevention or full nutrient repletion? Ann Intern Med 2015 Jan; 162(2):144-145.
Endocrine and Metabolism, Prevention/Screening.
Results provided by:
Copyright © 2016 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use
This PDF is available to Subscribers Only