Journal of Steroid Biochemistry & Molecular Biology 121 (2010) 334–337

Contents lists available at ScienceDirect

Journal of Steroid Biochemistry and Molecular Biology

journal homepage: www.elsevier.com/locate/jsbmb

Sun exposure questionnaire predicts circulating 25-hydroxyvitamin D

concentrations in Caucasian hospital workers in southern Italy夽
H.E.C. Hanwell a,∗ , R. Vieth a,b , D.E.C. Cole b , A. Scillitani c , S. Modoni c , V. Frusciante c ,
G. Ritrovato c , I. Chiodini d , S. Minisola e , V. Carnevale c
a
Department of Nutritional Sciences, University of Toronto, Department of Pathology and Laboratory Medicine, Mount Sinai Hospital, Toronto, Ontario, Canada M5G 1X5
b
Department of Laboratory Medicine & Pathobiology, University of Toronto, Toronto, Canada
c
Ospedale “Casa Sollievo della Sofferenza”, IRCCS, San Giovanni Rotondo (FG), Italy
d
Unit of Endocrinology and Diabetology, Ospedale Maggiore Policlinico IRCCS, Department of Medical Sciences, University of Milan, Italy
e
Department of Clinical Sciences, University of Rome “Sapienza”, Rome, Italy

a r t i c l e i n f o a b s t r a c t

Article history: Introduction: Recent sun exposure should correlate with circulating 25-hydroxyvitamin D [25(OH)D] due
Received 2 November 2009 to ultraviolet B (UVB)-catalyzed cutaneous synthesis of vitamin D.
Received in revised form 5 March 2010 Methods: A Sun Exposure Score was calculated for healthy adults using a recall questionnaire assessing
Accepted 8 March 2010
daily Time in Sun (<5 min, 5–30 min, >30 min) and Skin Exposure (face/hands; face/hands and arms;
face/hands and legs; and “bathing suit”) for 1 week in each of the winter and summer (n = 47 and 23,
Keywords:
respectively; n = 18 participated in both). Concentrations of 25(OH)D were measured by DiaSorin RIA on
Sun Exposure
end-of-week sera.
Vitamin D
UVB
Results: Mean serum 25(OH)D was higher in summer than winter (58.6 ± 16.5 nmol/L vs.
Questionnaire 38.8 ± 29.0 nmol/L, respectively, P = 0.003 unpaired). The calculated Sun Exposure Score correlated
Caucasian strongly with serum 25(OH)D during summer (Spearman’s rho = 0.59, P = 0.003); based on the Pearson
coefficient of determination, summer Sun Exposure Score explained 38% of the variability in summer
serum 25(OH)D. The Sun Exposure Score did not correlate with 25(OH)D in the winter (rho = 0.19,
P = 0.210). The summer correlation was largely explained by the Time in Sun (rho = 0.58, P = 0.004) rather
than area of Skin Exposed (rho = 0.10, P = 0.660). Although there was a correlation between winter and
summer Sun Exposure Scores (rho = 0.63, P = 0.005), there was no summer vs. winter correlation in serum
25(OH)D (rho = 0.08, P = 0.76).
Conclusion: This simple 1-week sun exposure recall questionnaire predicted summer serum 25(OH)D
concentrations, accounting for 38% of the variability in 25(OH)D among healthy Italian adults.
© 2010 Elsevier Ltd. All rights reserved.

1. Introduction The importance of sun exposure for vitamin D status is

highlighted by the observations from temperate regions where
In humans, vitamin D is primarily derived through the inter- circulating 25(OH)D concentrations fluctuate annually, lagging
action of the sun’s ultraviolet B (UVB) radiation with a cutaneous approximately 2 months behind the incident solar radiation [1–3].
cholesterol precursor, 7-dehydrocholesterol. Vitamin D from both Interestingly, serum 25(OH)D concentrations among those living
cutaneous synthesis and dietary or supplemental intake undergoes in temperate regions are commonly below the established range
hepatic hydroxylation to form 25-hydroxyvitamin D (25(OH)D), the for optimal vitamin D status [4], not only during winter and early
circulating concentrations of which are the objective biomarker of spring, but also throughout much of the year [2,5,6]. It is pos-
vitamin D status. A further renal hydroxylation produces the active sible that algorithms may be developed that will direct serum
metabolite, 1,25(OH)2 D. 25(OH)D analyses towards those at highest risk of impaired vita-
min D status. We hypothesize that people spending little time
outdoors during the day or exposing very little skin to the sun
夽 Special issue selected article from the 14th vitamin D Workshop held at Brugge, are at an elevated risk of having low serum 25(OH)D concentra-
Belgium on October 4–8, 2009. tions. The present study evaluates this hypothesis by evaluating
∗ Corresponding author at: Department of Nutritional Sciences, Faculty of the relationship between responses to a simple sun exposure recall
Medicine, University of Toronto, Pathology and Laboratory Medicine, Mount Sinai questionnaire and concurrent serum 25(OH)D concentrations dur-
Hospital, 600 University Avenue, Toronto, Ontario, Canada M5G 1X5.
ing the winter and summer seasons in healthy adults living in
Tel.: +1 416 586 4800x2726; fax: +1 416 586 8628.
E-mail address: heather.hanwell@gmail.com (H.E.C. Hanwell). Southern Italy.

0960-0760/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jsbmb.2010.03.023
 H.E.C. Hanwell et al. / Journal of Steroid Biochemistry & Molecular Biology 121 (2010) 334–337 335

2. Materials and methods Table 1

Participant characteristics at the winter and summer visits.

2.1. Participants Winter Summer P-Value

(n = 47) (n = 23)
Healthy adult Italian Caucasian hospital workers in southern Mean age (y)† 45.6 ± 13.5 42.2 ± 9.0 0.288
Italy (latitude: 40◦ N) volunteered for this study, providing signed Sex (males:females) 17:30 10:13 0.607
informed consent. Participants were eligible if physical examina- Body Mass Index (kg/m2 ) 25.5 ± 4.5 24.0 ± 3.4 0.126
Serum ionized calcium (mmol/L) 1.21 ± 0.03 1.24 ± 0.2 <0.001
tions and routine laboratory tests of renal and hepatic function
Vitamin D Status
were normal. Participants were excluded if they had conditions Mean serum 25(OH)D (nmol/L) 38.8 ± 29.0 58.6 ± 16.5 0.003
associated with impaired vitamin D metabolism or used medica- 25(OH)D < 75 nmol/L (n, %) 41 (87%) 18 (78%) 0.485
tion(s) known to affect calcium metabolism. The ethics committee 25(OH)D < 30 nmol/L (n, %) 25 (53%) 1 (4%) <0.001
of the “Casa Sollievo della Sofferenza” Hospital, San Giovanni †
Results presented as means ± SD or n participants with percent (in brackets).
Rotondo (FG), Italy, approved this study.

2.2. Data collection only in the lowest category (face and hands) during winter; hence
the uneven distribution in Fig. 2C.
The study was carried out during January and February 2003 In summer, the correlation between the weekly “Time in Sun”
(“winter” visit), and July and August 2003 (“summer” visit). Fast- score was almost as strongly correlated with serum 25(OH)D as the
ing blood and random urine samples were collected between weekly Total Sun Exposure Score (rho = 0.58, P = 0.004; Fig. 2B). In
07.00–09.00 h, and stored at −70 ◦ C until assayed as part of the contrast, the amount of skin exposed to the sun (Skin Exposure)
assessment of the biochemical parameters related to bone and min- did not correlate significantly with serum 25(OH)D (rho = 0.10,
eral metabolism. Serum 25(OH)D concentrations were analyzed P = 0.66).
by radioimmunoassay (RIA; DiaSorin, Stillwater, MN, USA) with A subgroup of 18 participants (9 males and 9 females) attended
intra- and inter-assay CVs <7.2% and <12%, respectively. Serum ion- both study visits. In these 18 participants, there was no significant
ized calcium (iCa) was assessed via ion-specific electrode (AVL LIST correlation between seasons for serum 25(OH)D concentrations
GmbH Medizintechnik, Graz, Austria). (rho = 0.08, P = 0.76). The weekly Sun Exposure Scores and Time
During the sample collection visit, participants’ recollection in Sun Scores were, however, correlated between the sum-
of daily sun exposure over the previous week was assessed via mer and winter (respectively, rho = 0.63, P = 0.005 and rho = 0.84,
a questionnaire administered during the sample collection visit P < 0.001).
(Fig. 1). There were three choices for the amount of time spent
outdoors each day (0 ≤5 min, 1 = 5–30 min, and 2 = ≥30 min) and 4. Discussion
four choices for clothing or skin exposure while outdoors (1 = face
and hands only; 2 = face, hands and arms; 3 = face, hands and legs; The present study demonstrates that a simple questionnaire
and 4 = “bathing suit”). A score to estimate of their mean weekly assessing both the amount of time spent outdoors and amount of
sun exposure was calculated: The product of the amount of time skin exposed in a given week for healthy Caucasian adults living at
spent outdoors and the amount of skin exposed was calculated for latitude 40◦ N offers a good prediction of 25(OH)D concentrations in
each day to create a daily Sun Exposure Score (min = 0, max = 8). All summer. This questionnaire involves minimal participant burden
seven days’ Sun Exposure Scores were summed to equal the weekly but includes a timeframe long enough to estimate typical activity
Sun Exposure Score (min = 0, max = 56). Similarly, the weekly Time for an individual. This questionnaire could serve as a screening tool
in Sun Score and weekly Skin Exposure Score were each calculated to identify patients at increased risk of low sun exposure who might
by summing the seven daily scores (respectively, min = 0, max = 14 benefit from a serum 25(OH)D determination.
and min = 7, max = 28). The strength of the relationship detected between sun expo-
sure in the summer and serum 25(OH)D, highlights the key role
2.3. Statistics of summer sun exposure in determining vitamin D status in a Cau-
casian population at 40◦ N latitude. The present study did not collect
All statistical analyses were carried out using SPSS 13.0 (SPSS information regarding type of clothing worn, use of sunscreens, or
Inc., Chicago, IL). Means and standard deviations describe paramet- vacation travel, nor was dietary or supplemental intake assessed
ric data; medians and interquartile ranges describe non-parametric to quantify exogenous sources of vitamin D. While the relation-
data. The Pearson coefficient of determination and Spearman rho ship might have been improved by inclusion of such additional
describe relationships between the weekly Scores and 25(OH)D, measures, it also would have increased the complexity of the ques-
separately for each season. Paired t-tests (for comparisons of tionnaire, making routine clinical use less feasible and recall less
parametric variables between winter and summer for the n = 18 accurate. Furthermore, skin pigmentation affects the rate of cuta-
participants completing both visits), unpaired t-tests (for compar- neous vitamin D synthesis, and thus, the relationship between the
isons of all participants, n = 43 in winter and n = 27 in summer) and Sun Exposure Score and circulating 25(OH)D may vary with skin
the Fisher exact test were used to evaluate categorical variables. pigmentation. Since all participants were native Italians, future
P < 0.05 was set as the limit of significance. use of this questionnaire in ethnically diverse populations should
also objectively query participant skin pigmentation to evaluate its
3. Results potential influence on the relationship between sun exposure and
serum 25(OH)D.
Participant characteristics are presented in Table 1. Mean A comparable study of men working outdoors in Nebraska USA
serum 25(OH)D concentrations were higher in summer than focused more intently on the type and quality of clothing in addi-
winter (respectively, 58.6 ± 16.5 and 38.8 ± 29.0 nmol/L, P = 0.003 tion to time spent outdoors [7]. Despite more precise description
unpaired). The correlation between the serum 25(OH)D and weekly of Skin Exposure in that study, the correlation between sun expo-
Sun Exposure Score was significant in summer but not winter sure and serum 25(OH)D concentrations was weaker than in the
(respectively, Spearman rho = 0.59 and 0.19, P = 0.003 and 0.212; present study. The stronger summertime correlation observed in
Fig. 2A). Interestingly, all 47 participants reported Skin Exposure the present study may be partly explained by the relative eth-
336 H.E.C. Hanwell et al. / Journal of Steroid Biochemistry & Molecular Biology 121 (2010) 334–337

Fig. 1. Scoring for the weekly sun exposure recall questionnaire. Shown in gray are the values ascribed to each category and the method for deriving the Sun Exposure Score
(sum of the daily products of Time Outdoors and Skin Exposure). The range of the Sun Exposure Score is from the 0 (lowest amount of time spent outdoors and lowest amount
of skin exposed) to a maximum score 56 (outdoors for more than 30 min in a bathing suit every day).

nic and cultural homogeneity of the Italian participants. Also, the individuals with arms and legs exposed to large-angle incident
present correlation was almost entirely driven by Time in Sun, not sunshine [8]. The summer Skin Exposure score did not correlate
the amount of exposed skin, suggesting that the categories for time significantly with summer 25(OH)D concentrations. One possible
spent in the sun used in this study (<5 min, 5–30 min, >30 min) are explanation for this finding is that participants with moderately
easily recalled and likely to be of biological significance. The time- high Skin Exposure when in the sun did not report spending much
frames were selected based upon unpublished preliminary study time in the sun. People who spend very little time in the sun may
in this population and the 30-min time frame used for the high- be expected to have low 25(OH)D; if these participants wear light
est category is only slightly longer than the estimated time needed clothing (arms and/or legs uncovered) most days but are outdoors
to achieve maximal dermal synthesis of vitamin D in fair skinned less than 5 min each day of the week, their 25(OH)D status would

Fig. 2. Serum 25(OH)D concentrations in summer (black open circles) and winter (small grey circles) as a function of Sun Exposure Score (A) derived from one-week recall
of Time in Sun (B) and Skin Exposure (C). Sun exposure was determined from a questionnaire scoring for each of the preceding 7 days as follows: Time in Sun: <5 min = 0;
≥5 min but ≤30 min = 1; and ≥30 min. Skin Exposure: Face and hands only = 1; Face and arms = 2; Face, hands and legs = 3; “Bathing suit” = 4. Each day’s Time in Sun and Skin
Exposure results were multiplied; the sum of all 7 days resulted in derivation of the weekly Sun Exposure Score. Each panel displays the regression line (with regression
equation, r2 and P-value) and the 95% Confidence Interval for the summertime data.
 H.E.C. Hanwell et al. / Journal of Steroid Biochemistry & Molecular Biology 121 (2010) 334–337 337

be expected to correspond with their low Time in Sun Scores and findings emphasize the importance of outdoor activity as a way
low overall Sun Exposure Scores but would not correspond with to maintain adequate vitamin D nutritional status. In spite of its
their high Skin Exposure Scores. It is in these sorts of instances that limitations, the present study offers a useful tool for clinicians to
the importance of querying both Time in Sun and Skin Exposure administer to patients to obtain a proxy measure of 25(OH)D mea-
becomes obvious, as the derived Sun Exposure Score reflects actual surements.
sun exposure habits more accurately than querying either variable
alone. References
In contrast to the summer data, the Sun Exposure Score did not
correlate with serum 25(OH)D in the winter. Given that there is [1] V. Carnevale, S. Modoni, M. Pileri, et al., Longitudinal evaluation of vitamin D
status in healthy subjects from southern Italy: seasonal and gender differences,
insufficient solar UVB to generate vitamin D in the skin at latitudes Osteoporos. Int. 12 (2001) 1026–1030.
>40◦ N during the winter months [8], the participants would not [2] P.B. Rapuri, H.K. Kinyamu, J.C. Gallagher, V. Haynatzka, Seasonal changes in cal-
produce vitamin D during time spent outdoors at that time. It would ciotropic hormones, bone markers, and bone mineral density in elderly women,
J. Clin. Endocrinol. Metab. 87 (2002) 2024–2032.
be of interest to investigate the strength of the relationship between [3] R. Vieth, D.E. Cole, G.A. Hawker, H.M. Trang, L.A. Rubin, Wintertime vitamin D
winter sun exposure and serum 25(OH)D at varying latitudes and insufficiency is common in young Canadian women, and their vitamin D intake
explore this relationship in other populations of varied ethnicities does not prevent it, Eur. J. Clin. Nutr. 55 (2001) 1091–1097.
[4] B. Dawson-Hughes, R.P. Heaney, M.F. Holick, P. Lips, P.J. Meunier, R. Vieth, Esti-
with cultural practices that may affect the relationship between
mates of optimal vitamin D status, Osteoporos. Int. 16 (7) (2005) 713–716.
sun exposure and serum 25(OH)D concentrations. [5] S. Harris, B. Dawson-Hughes, Seasonal changes in plasma 25-hydroxyvitamin D
Winter sun exposure habits, particularly time spent in the sun, concentrations of young American black and white women, Am. J. Clin. Nutr. 67
strongly reflected the summer sun exposure habits. However, in the (1998) 1232–1236.
[6] J.D. Maxwell, Seasonal variation in vitamin D, Proc. Nutr. Soc. 53 (1994) 533–543.
winter, all subjects indicated that they exposed only their face and [7] M.J. Barger-Lux, R.P. Heaney, Effects of above average summer sun exposure on
hands, resulting in identical responses for the Skin Exposure vari- serum 25-hydroxyvitamin D and calcium absorption, J. Clin. Endocrinol. Metab.
able. This finding limited statistical analyses but also highlighted 87 (2002) 4952–4956.
[8] A.R. Webb, L. Kline, M.F. Holick, Influence of season and latitude on the cutaneous
the uniformly minimal wintertime Skin Exposure in this group of synthesis of vitamin D3: exposure to winter sunlight in Boston and Edmonton
Italian adults. will not promote vitamin D3 synthesis in human skin, J. Clin. Endocrinol. Metab.
While the present observations may apply only to a Caucasian 67 (1988) 373–378.
population (and need to be confirmed in other populations), these