Calcium for health


In by Raphikammer

Calcium is an alkaline earth, divalent, cationic element,

abundant in the biosphere, and widely distributed in nature.


It exhibits intermediate solubility. As a solid, calcium

forms crystalline minerals with various anions. These salts

make up the bulk of limestone, marble, gypsum, coral,

pearls, seashells, bones, and antlers. In solution, the calcium

ionic radius (0.99 ˚A units) allows the ion to fit snugly

into the folds of protein molecules.


Calcium is unusual—perhaps unique—among the nutrients

in that its intake (whether from foods or supplements)

is not related to its primary intracellular, metabolic

function. Rather, calcium nutrition is centered almost

exclusively on the secondary functions of the nutrient.

Accordingly, the primary functions are described here for

completeness, but only briefly. More information can be

found in standard textbooks of cell physiology or in reviews

of calcium signaling (1).


Primary Metabolic Functions

Calcium acts as a second messenger within cells, linking

external stimuli acting on cells to the specific, internal

responses a cell is able to make (e.g., nerve signals and

muscle contraction). By forming up to 8 to 12 coordination

bonds with oxygen atoms in amino acid side chains,

calcium stabilizes the tertiary structure of numerous catalytic

and structural proteins. Cytosolic calcium ion levels

are normally maintained at very low concentrations

[3–4 orders of magnitude below extracellular fluid (ECF)

levels]. The second messenger response occurs when calcium

ions flood into critical cytosolic compartments in

response to first message stimuli.

Additionally, dissolved calcium in the circulating

blood and ECF of all vertebrates supports such diverse

functions as blood clotting and neuromuscular signal

transmission. Calcium is not consumed in the exercise of

these metabolic functions.

ECF [Ca2+] is tightly maintained at approximately

4.4 to 5.2 mg/dL (1.1–1.3 mmol/L). The regulatory apparatus

behind this constancy consists of parathyroid

hormone (PTH), calcitonin, and 1,25-dihydroxyvitamin D

[1,25(OH)2D], acting jointly through control of intestinal

calcium absorption efficiency, bone resorption, and the renal

excretory threshold for calcium.

Secondary Functions

Effects on the Size and Strength of the Nutrient Reserve

(Bone Mass)

Calcium is lost continuously from the body through shed

skin, hair, nails, sweat, and excreta. For this reason, landliving

vertebrates, needing a continuous supply of calcium,

have evolved an internal reserve, in the form of

bone. Because bone also serves structural/mechanical

functions, the reserve has become far larger than would be

needed solely to protect calcium’s primary functions. It is

for this reason that the primary functions themselves are

not threatened by deficient calcium intake, or enhanced

by calcium repletion.

The bony reserves are accessed by a process termed

“bone remodeling.” Bony tissue is continuously renewed

by first resorbing preexisting volumes of bone and then

subsequently replacing them with new bone. Mineralization

of the new bone occurs at a rate that is the integral of

the prior several days of osteoblast activity, and for that

reason tends to be relatively constant over the short term.

By contrast, osteoclastic bone resorption is controllable

minute by minute. Thus, by modulating bone resorption,

the body can, in effect, withdraw calcium from, or cause

it to be taken up by, bone whenever ECF [Ca2+] departs

from optimal levels.

When daily absorbed calcium intake is less than that

needed to offset daily calcium losses, bone resorption exceeds

bone formation and the bony reserves are depleted.

This occurs by net destruction of microscopic volumes

of bony tissue and scavenging of the calcium released in

the process. Such decrease in skeletal mass results in a

corresponding reduction in strength. Additionally, bone

remodeling itself directly contributes to bony structural

weakness (2), insofar as the remodeling locus is, for the

several months of its life cycle, depleted of its normal

complement of bony material, thereby greatly weakening

the involved microscopic bony elements.

The principal purpose of calcium intake during

growth is to support the accumulation of the skeletal

mass called for in the genetic program, that is, the building

of a large calcium reserve. During the adult years,

intake serves to (i) offset daily losses, thus preventing

unbalanced withdrawals from the skeletal reserves, with

their inevitable, associated reduction in bony strength; and

(ii) reduce the level of bone remodeling to the minimum

needed for optimum structural maintenance (2). These

two effects are the basis for the protective effect of calcium

with respect to osteoporosis.

Intraluminal Effects of Unabsorbed Dietary Calcium

Net absorption efficiency for ingested calcium is of the order

of 10% to 15% (see later). Accordingly, up to 90% of dietary

and supplemental calcium remains in the intestinal

lumen and is excreted as a component of the feces. At high

calcium intakes, unabsorbed calcium amounts to 1000 mg

(25 mmol) per day or more. This unabsorbed calcium complexes

with other constituents of the digestive residue,

blocking their absorption or neutralizing their luminal actions

(3). This occurs, for example, with oxalic acid, which

may be either present in ingested plant foods or produced

by bacterial degradation of unabsorbed food fatty acids.

The formation of calcium oxalate in the gut lumen reduces

oxalate absorption and hence the renal oxalate load. It

thereby reduces the risk of kidney stones. Similarly, the

calcium ion complexes directly with free fatty acids and

bile acids in the digestate substances, which, in their free

form, act as mucosal irritants. In colon cancer–prone individuals,

these irritants would otherwise serve as cancer

promoters. These intraluminal actions are the basis for the

protective effects of high calcium intakes on risk of renal

stone disease and colon cancer.

Additionally, calcium complexes with dietary phosphorus,

blocking its absorption to some extent. This is the

basis for the use of calcium salts as a part of the control of

hyperphosphatemia in patients with end-stage renal disease

(ESRD). Every 500 mg of ingested calcium (whether

from foods or supplements) binds ≈166 mg of coingested

phosphorus, preventing its absorption (4).

“Off-Loop” Effects of Alterations in Calcium Homeostasis

When calcium intake is low, PTH is secreted to improve

renal calcium conservation and intestinal absorption efficiency,

the latter through 1–hydroxylation of 25(OH)D

to 1,25(OH)2D in the kidney. The calcium-conserving effects

of these hormones are part of a classical negative

feedback loop, in the sense that 1,25(OH)2D, by increasing

calcium absorptive extraction from food, counteracts

to some extent the original stimulus to PTH secretion and

1,25(OH)2D synthesis.

In addition to these functions within the feedback

control loop, 1,25(OH)2D binds to membrane receptors

in many tissues not directly involved in calcium regulation

(3). These include vascular smooth muscle cells and

adipocytes. These effects are termed “off-loop,” because

they occur as a result of reduced ECF [Ca2+] but do not

act to change that level. Hence, they do not influence the

signals that caused them in the first place, that is, they

are not a part of the regulatory feedback loop. The cell

membrane receptors are linked to calcium channels that

open and let calcium ions into the cytosol, where they

may trigger their usual second messenger function (but

without the normal first messenger). The presence of high

cytosolic calcium levels, when dietary calcium is low, has

given rise to the term “calcium paradox disease.” In individuals

with limited control of cytosolic [Ca2+], this rise

in cytosolic calcium triggers inappropriate, tissue-specific

cell activity, for example, smooth muscle contraction in

arterioles and adipogenesis in fat cells. These relationships

are the basis for the protective effects of high calcium intake

against hypertension and obesity, and probably for

premenstrual syndrome and polycystic ovary syndrome

as well (3).


The Internal Calcium Economy

The adult human body contains approximately 1000 to

1300 g (25,000–32,500 mmol) of calcium, with more than

99% being locked up in bones and teeth. Low hydration

of bone, together with the insolubility of hydroxyapatite

(the principal form of calcium phosphate in mineralized

tissues), means that most body calcium is effectively exterior

to the ECF and accessible only by cellular action (e.g.,

osteoclastic bone resorption).

The ECF, which is the locus of all body calcium traffic,

contains about 1 g (25 mmol) calcium (i.e., ≈0.1% of

total body calcium). Soft tissues contain another 7 to 8 g

(175–200 mmol) of calcium, mostly locked up in intracellular

vesicles, which store calcium for its critical, second

messenger function. The calcium homeostatic regulatory

apparatus functions solely to maintain the constancy of

the concentration of the ≈1 g of calcium in the ECF. In

healthy midlife adults, ECF calcium turns over at a rate

of approximately 650 mg/day (≈10 mg/kg/day), with

bone mineralization and resorption accounting for half to

two-thirds of that traffic.


The calcium economy of a middle-aged woman. In considering

the magnitudes of these transfers, it is important to

recognize that they do not vary independent of one another.

An increase in absorption, for example, produces

an immediate decrease in bone resorption. This linkage

is mediated by the PTH–calcitonin–vitamin D regulatory



Calcium is absorbed mainly from the small intestine by a

combination of active, transcellular transport and passive,

paracellular diffusion.

The active transport component

is mediated by a vitamin D–dependent calcium-binding

protein (“calbindin”) that shuttles calcium ions from

the luminal brush border to basolateral portions of the

cell membrane, where calcium is released into the ECF.

Calbindin activity is highest in the duodenum and drops

along the length of the remaining bowel (including the

colon). Accordingly active transport capacity is greatest

in the duodenum. However, the residence time of the

digestate in the duodenum is short, and most of the actual

mass transport occurs in the jejunum and ileum, where

residence time is longer.

The partition of absorption between the active and

the passive mechanisms is not well studied, but data

from various sources suggest that, at nutritionally relevant

intake loads (i.e., 7.5+ mmol/meal), passive absorption

amounts to approximately 15% of intake. Fractional

absorption above that value thus reflects the vitamin D–

mediated active transport component. The latter is highly

variable, both because it is physiologically regulated in

response to body need for calcium, and, in part, because it

is often limited by vitamin D availability. The interactions

between intake load of calcium and its active absorption

are complex and are summarized in Figure 2. As measured

fractional absorption is typically on the order of

0.30 to 0.32; it follows that the active transport component

amounts to approximately 0.16 (16%). As is shown

in Figure 2, the 16% isogram intersects the dashed line for

5 mmol (200 mg) net absorption at an intake of approximately

1200 mg (30 mmol), and thus designates the oral

intake needed to maintain total body equilibrium. When

vitamin D status is less than optimal, the body is generally

unable to maintain active absorption at a 16% level

and absorption occurs along lower and lower isograms

until, at severe vitamin D deficiency, active absorption is

zero. As is shown in Figure 2, intake at such absorption

values would need to be in the range of 3000 mg/day to

ensure absorption of sufficient calcium to offset obligatory


the calcemic rise above baseline in healthy adults for a 500-mg calcium supplement source ingested as part of a low-calcium breakfast. It illustrates

a number of features of calcium absorption: (i) a delay of

approximately 30 minutes before serum calcium begins to

rise, reflecting gastric residence time; (ii) peak calcemia at

3 to 5 hours after ingestion, indicating continuing absorptive

input throughout that period of time; (iii) a degree of

calcemia approximating a 1% rise for every 100 mg calcium

ingested, that is, a perturbation that is within the

usual normal range for serum calcium and hence effectively

undetectable outside of a research context; and (iv)

gradual return to baseline by 9 to 10 hours. Tracer studies

show that calcium absorption is effectively completed by

five hours after ingestion (5), and the slow fall to baseline

after the peak reflects offsetting declines in other inputs

into the ECF.

It is commonly considered that calcium salts must

be dissociated to be absorbed, and hence that solubility

predicts absorbability. However, this is probably incorrect.

The pH of the digestate in the small intestine is close

to neutral, and it is likely that most of the digestate calcium

is complexed with prevailing anions in the digestate.


Relationship of vitamin D–mediated, active calcium

absorption, calcium intake, and net calcium gain across the gut.

Each of the contours represents a different level of active absorption

above a baseline passive absorption of 12.5%. (The values

along each contour represent the sum total of passive and variable

active absorption.) The horizontal dashed lines indicate 0

and 5 mmol/day net absorption, respectively. The former is the

value at which the gut switches from a net excretory to a net absorptive

mode, and the latter is the value needed to offset typical

urinary and cutaneous losses in mature adults.


Time course of the rise in serum calcium following a single oral

dose of a commercial calcium carbonate preparation (containing 500 mg

calcium) taken as part of a light, low-calcium breakfast. Error bars are 1 SEM.

Source: Copyright Robert P. Heaney, 2001, 2004; used with permission.

Aqueous solubility of calcium salts spanning 4 to 5 orders

of magnitude has been shown to have little or no effect on

absorbability if the calcium source is coingested with food

(6). Double-tracer studies have demonstrated absorption

of insoluble calcium complexes without prior dissociation

(7). Thorough dispersion of calcium salts among food particulates

is probably more important than actual solubilization.

Additionally, continuous slow release of calcium

from the stomach, exposing the duodenal mucosa to only

small amounts of calcium at a time, substantially improves



Calcium leaves the body through unabsorbed digestive

secretions, through sweat and shed skin, hair, and nails,

and through urine

In non exercising adult humans

with typical calcium intakes, digestive calcium losses

amount to approximately 120 mg/day, cutaneous losses

amount to approximately 60 mg/day, and urinary

losses amount to approximately 120 mg/day, with

great individual variability around these figures. Only

the urinary loss is physiologically regulated by the

system controlling calcium homeostasis, and much of

even the urinary calcium represents obligatory loss,

that is, excretion determined by forces outside of the

calcium regulatory system (9), such as salt intake and

net endogenous acid production (as, for example, from

metabolism of S-containing amino acids). On average,

urine calcium rises by approximately 45 mg (1.1 mmol)

for every 1000 mg (25 mmol) increase in calcium intake.

This increase is a reflection of the small absorptive

calcemia (Fig. 3), which produces a corresponding rise in

the filtered load of calcium.

In adults, the primary purpose served by ingested

calcium is the offsetting of obligatory excretory losses,

thus protecting the skeletal reserves and thereby preserving

their structural integrity. Ingested calcium, thus, does

not so much “go” to bone as prevents net removal of calcium

from bone.



Calcium is a nutrient and would normally be ingested as

a component of food. However, except for dairy foods,

modern diets, especially seed-based plant foods (which

are the basis of most contemporary diets), are calciumpoor

diets. Hence, for many individuals, achieving an adequate

calcium intake may be difficult without recourse

to supplements or calcium-fortified foods. (The latter are

effectively equivalent to taking a supplement along with

the otherwise unfortified food.)


Supplementation to Achieve Recommended

Intake Levels

Diets free of dairy foods typically contain no more than

200 to 300 mg calcium, far below currently recommended

intakes (Table 1). Supplementation (or fortification) will

often be required to meet optimal intake objectives.

Table 1 Estimated Average Requirements (EARs) for Calcium and the

Corresponding RDAs (mg/day)

Age range EAR RDAa

Infants, 7–12 mo 270 350


1–3 yr 500 600

4–8 yr 800 1000

Boys and girls, 9–18 yr 1300 1550

Men and women

19–50 yr 1000 1200

>50 yr 1200 1450

aUsing an estimated 10% coefficient of variation of individual requirements

around the population mean.

Source: From Ref. 10.

Calcium 105

Since absorption efficiency is inversely proportional

to the logarithm of the ingested load (11), absorption is

maximized by a divided dose regimen (e.g., 3°ø per day;

Fig. 4). Also, because delivery of calcium to the absorptive

sites in the upper small intestine is optimized under meal

conditions, it is best to take calcium supplements with

meals. (N.B.: Fortified foods tend, automatically, to meet

both objectives.)


The nutritional preparations of calcium include mainly

salts with such anions as carbonate, citrate, phosphate,

lactate, acetate, fumarate, and citrate-malate (CCM). In

addition, salts with gluconic acid may occasionally be

found, and calcium chelates with amino acids are also

marketed. The calcium content (i.e., “elemental” calcium)

varies from 40% for the carbonate salt to ≈13% for CCM.

For phosphate binding in ESRD, the acetate salt is more

commonly used. In the United States, most preparations

come in the form of swallowable or chewable tablets, with

calcium contents ranging from 200 to 600 mg per tablet.

Bioavailability is approximately the same for all the

leading salts, although CCM and the chelates tend toward

the high end of the range and the gluconic acid salts toward

the low end. Absorbability of the salt is only very

weakly related to solubility, and gastric acid is not necessary

for calcium absorption if the supplement is taken

(as recommended) with meals. The most extensive, sideby-

side comparisons have involved the carbonate and citrate

salts, and the bulk of the evidence for such comparisons

indicates approximately equal absorbability for the

two sources, with perhaps a slight edge for the carbonate

salt (12). Poor pharmaceutical formulation will impede

disintegration and hence impair absorption, a problem

encountered with many generic calcium supplement

products sold in the 1980s and early 1990s (13). For this

reason, preference should be given to supplements that

meet United States Pharmacopeia (USP) disintegration

standards, and, even better, to those that have demonstrated


A growing variety of fortified foods has been available

since 1999. As noted, fortification tends to improve

the nutritional value of low-calcium foods and, to some

extent, it can be thought of as equivalent to taking supplements

with meals. However, interactions between added

calcium and various food constituents during food processing

and storage may alter the absorbability characteristics

of the former. For example, it was noted during the

early days of juice supplementation that CCM was well

absorbed from orange and grapefruit juices, and even better

from apple juice, but poorly from lemon juice. These

differences could not have been predicted from what was

known of food chemistry. Hence, with fortified foods as

with supplements, actual bioavailability of the product

reaching the consumer should be demonstrated.

When calcium is added to beverages (such as orange

juice or soy beverage), an additional problem arises. Solubility

of the principal calcium salts is relatively low, and

serving size portions of such beverages would not sustain

in solution more than a small fraction of the calcium

content of, say, a comparable serving of milk. Hence, such

fortification almost always requires physical suspension

of a particulate. In some beverages, this suspension is so

poor that the calcium settles as a dense sludge at the bottom

of the beverage container and may, accordingly, not

be ingested at all (14).

Supportive Therapy as a Part of Anti Osteoporosis


Current anti osteoporosis pharmacotherapy includes bisphosphonates,

selective estrogen receptor modulators

(SERMs), estrogen, and anabolic agents such as the fluoride

salts, PTH derivatives such as teriparatide, and

RANK-ligand antibodies and cathepsin-K inhibitors. All

have at least stabilization of bone mass as their goals.

Some of them, such as the bisphosphonates, can lead

to slow, steady-state bone gain (0.5–1.0% per year), and

the anabolic agents can produce as much as 8% to 10%

bone gain per year. To support this increase, especially for

the anabolic agents, calcium intake from diet must usually

be augmented by supplements. Optimal doses for calcium

during pharmacotherapy have not been established.However,

all the bisphosphonates and SERMs have been tested

only with 500 to 1000 mg supplemental calcium, whereas

fluoride has been shown to produce bone hunger calling

for as much as 2500 mg of calcium per day. Only estrogen

has been studied with and without supplemental calcium,

and here the evidence is very clear: bony effects of estrogen

are augmented two- to threefold, and estrogen dose

can be reduced by half if calcium intake is above 1000

mg/day (15,16). With the more potent anabolic agents, a

calcium phosphate preparation may be preferable, so as

to ensure an adequate intake of both of the components of

bone mineral and to compensate for the intestinal binding

of diet phosphorus by high-dose calcium supplementation.

The high phosphorus loads of the phosphate salts

produce no adverse metabolic consequences, and calcium

phosphate supports bone anabolism fully as well as the

carbonate salts.


Ancillary Therapy for Prevention or Treatment of Miscellaneous Disorders

Hypertension, pre-eclampsia, colon cancer, renolithiasis,

premenstrual syndrome, polycystic ovary syndrome,

and obesity—all multifactorial disorders—have each been

shown to have a calcium-related component (3), and

for several of them, calcium supplementation has been

shown in randomized-controlled trials to reduce incidence

and/or severity. Optimal calcium intake for this protection

has not been established for any of the disorders

concerned, but several threads of evidence indicate that

total intakes of 1200 to 1800 mg of calcium per day may

be sufficient. The role of calcium in these disorders has

been described earlier (see sections “intraluminal effects of

unabsorbed dietary calcium” and “off-loop effects of alterations

in calcium homeostasis”). Specific pharmacotherapy

of any of the disorders concerned should always be

accompanied by an adequate calcium intake, using supplements

if necessary.



There are few, if any, true contraindications to calcium

supplementation. In general, supplementation moves

106 Heaney

contemporary intakes into the range that would have been

the Paleolithic standard, and hence helps to normalize

modern diets. However, patients receiving calcitriol therapy

or suffering from disorders such as sarcoidosis, in

which calcium absorption may be high, should not take

supplements except under medical supervision.



Calcium supplements may bind with tetracycline antibiotics

and hence reduce their absorbability. The element has

also been reported to interfere slightly with thyroxin absorption.

Hence, a person requiring both calcium and thyroid

replacements should take them at different times of

the day or have plasma thyroxin and thyroid-stimulating

hormone (TSH) levels checked to ensure that the thyroid

dose produces the desired therapeutic effect. Both calcium

salts and high-calcium foods reduce absorption of nonheme

iron ingested at the same meal in unprepared subjects.

However, chronic supplementation studies show no

long-term deterioration in iron status in adults and no

interference with augmenting iron status during growth

(17). The single-meal tests that are used to demonstrate

this interference could not have detected physiological

upregulation of iron absorption.

Adverse reactions tend to be extremely rare and

mostly idiosyncratic. Although constipation is often said

to be a consequence of taking calcium carbonate, the evidence

is scant (18), and in several randomized-controlled

trials, the difference in degree of constipation between the

calcium- and placebo-treated groups has generally been

small and usually not statistically significant.



The Food and Nutrition Board of the Institute of Medicine,

in its 1997 recommendations, set a tolerable upper intake

level for calcium to be 2500 mg/day (9). However, it is important

to note that there has never been a reported case of

overdose of calcium from food sources, even at continuing

intakes over 6000 mg/day. Supplement intakes above

2500 mg/day are occasionally associated with a syndrome

similar to the milk alkali syndrome. The pathogenesis of

the hypercalcemia seen in this condition is complex, but

there is usually hypoperfusion of both the kidneys and the

skeleton, the two most important internal regulatory organs

for calcium. The condition can usually be managed

by giving attention to adequate hydration and maintenance

of blood flow to these critical organs. Except as support

for the most potent osteoporosis pharmacotherapy, or

in management of the hyperphosphatemia of ESRD, there

is no known reason to use supplements at a dose above

2500 mg of calcium per day.



Calcium supplements are regulated as foods in the United

States. Bioavailability is not a regulated characteristic of

marketed supplement products. Nevertheless, because of

pharmaceutical formulation and food matrix effects on

absorbability, bioavailability of different preparations of

the same salt (e.g., calcium carbonate) may vary over a

twofold range.



1. Awumey EM, Bukoski RD. Cellular functions and fluxes of

calcium. Calcium and Human Health. Totowa, NJ: Humana,


2. Heaney RP. Is the paradigm shifting? Bone 2003; 33:457–


3. Heaney RP. Ethnicity, bone status, and the calcium requirement.

Nutr Res 2002; 22(1–2):153–178.

4. Heaney RP, Nordin BEC. Calcium effects on phosphorus absorption:

Implications for the prevention and co-therapy of

osteoporosis. J Am Coll Nutr 2002; 21(3):239–244.

5. Barger-Lux MJ, Heaney RP, Recker RR. Time course of calcium

absorption in humans: Evidence for a colonic component.

Calcif Tissue Int 1989; 44:308–311.

6. Heaney RP, Recker RR,Weaver CM. Absorbability of calcium

sources: The limited role of solubility. Calcif Tissue Int 1990;


7. Hanes DA, Weaver CM, Heaney RP, et al. Absorption of

calcium oxalate does not require dissociation in rats. J Nutr

1999; 129:170–173.

8. Heaney RP, Berner B, Louie-Helm J. Dosing regimen for

calcium supplementation. J Bone Miner Res 2000; 15(11):


9. Nordin BEC, Polley KJ, Need AG, et al. The problem

of calcium requirement. Am J Clin Nutr 1987; 45:1295–


10. Food and Nutrition Board, Institute of Medicine. Dietary

Reference Intakes for Calcium, Magnesium, Phosphorus, Vitamin

D, and Fluoride.Washington, D.C.: National Academy

Press, 1997.

11. Heaney RP, Weaver CM, Fitzsimmons ML. The influence of

calcium load on absorption fraction. J Bone Miner Res 1990;


12. Heaney RP, Dowell MS, Barger-Lux MJ. Absorption of calcium

as the carbonate and citrate salts, with some observations

on method. Osteoporos Int 1999; 9:19–23.

13. Shangraw RF. Factors to consider in the selection of a calcium

supplement. In: Proceedings of the 1987 Special Topic Conference

on Osteoporosis. Public Health Report 1989; S104:


14. Heaney RP, Rafferty K, Bierman J. Not all calcium-fortified

beverages are equal. Nutr Today 2005; 40:39–44.

15. Nieves JW, Komar L, Cosman F, et al. Calcium potentiates

the effect of estrogen and calcitonin on bone mass: Review

and analysis. Am J Clin Nutr 1998; 67:18–24.

16. Recker RR, Davies KM, Dowd RM, et al. The effect of

low dose continuous estrogen and progesterone therapy

with calcium and vitamin D on bone in elderly women:

A randomized controlled trial. Ann Intern Med 1999; 130:


17. Ilich-Ernst JZ, McKenna AA, Badenhop NE, et al. Iron status,

menarche, and calcium supplementation in adolescent girls.

Am J Clin Nutr 1998; 68:880–887.

18. Clemens JD, Feinstein AR. Calcium carbonate and constipation:

A historical review of medical mythopoeia. Gastroenterology

1977; 72:957–961.