Arch Osteoporos (2014) 9:197 DOI 10.1007/s11657-014-0197-9
ORIGINAL ARTICLE
Vitamin D response of older people in residential aged care to sunlight-derived ultraviolet radiation Seeta Durvasula & Peter Gies & Rebecca S. Mason & Jian Sheng Chen & Stuart Henderson & Markus J. Seibel & Philip N. Sambrook & Lynette M. March & Stephen R. Lord & Cindy Kok & Monique Macara & Trevor R. Parmenter & Ian D. Cameron
Received: 28 February 2014 / Accepted: 24 September 2014 / Published online: 14 October 2014 # International Osteoporosis Foundation and National Osteoporosis Foundation 2014
S. R. Lord Neuroscience Research Australia, University of NSW, Randwick, NSW 2031, Australia
Purpose The purpose of this study is to measure the ultraviolet radiation (UVR) exposure of those in residential aged care in an earlier trial of sunlight exposure and to determine its effect on their vitamin D response. Methods Attendance data, demographic, clinical and biochemical variables for 248 participants were used for a secondary analysis of a previous cluster randomized trial of sunlight exposure and falls. The ambient solar UV Index data were used to calculate the participants’ UVR dose. Multiple linear regression was used to test if UVR exposure over 6 months, as measured by the standard erythemal dose (SED), was a predictor of vitamin D response, controlling for age, gender, BMI, calcium intake, baseline vitamin D and season of exposure. Results The median 25-hydroxyvitamin D (25OHD) was 32.4 nmol/L at baseline and 34.6 nmol/L at 6 months (p=0.35). The significant predictors of 25OHD at 6 months were UVR exposures in spring-summer (coefficient=0.105, 95 % confidence interval (CI) 0.001–0.209, p=0.05) and autumn-winter (coefficient=0.056, 95 % CI 0.005–0.107, p = 0.03) and baseline vitamin D (adjusted coefficient= 0.594, 95 % CI 0.465–0.724, p=0.00). In those starting sunlight sessions in spring, an increase of 1 unit in log SED was associated with 11 % increase in 25OHD. Conclusions Natural UVR exposure can increase 25OHD levels in older people in residential care, but depends on the season of exposure. However, due to inadequate sun exposure, 25OHD did not reach optimal levels. Nevertheless, where sun exposure is encouraged in this group, the focus for the start of exposure should be in the months of spring or autumn, as this timing was associated with a vitamin D response.
I. D. Cameron The John Walsh Centre for Rehabilitation Research, Sydney Medical School Northern, University of Sydney, St Leonards, NSW 2065, Australia
Keywords Sunlight . Ultraviolet radiation . Older people . Residential aged care . Vitamin D
Abstract Summary The aim of this study was to determine the vitamin D response to sunlight ultraviolet radiation in older people. Increases in vitamin D depended on the season of exposure, but the changes were small. Natural sun exposure is not a practical intervention for vitamin D deficiency in this population.
Philip N. Sambrook passed away during the course of this study. S. Durvasula (*) : T. R. Parmenter Sydney Medical School Northern, University of Sydney, Sydney, NSW 2006, Australia e-mail: [email protected] P. Gies : S. Henderson Australian Radiation Protection and Nuclear Safety Agency, 619 Lower Plenty Road, Yallambie, VIC 3085, Australia R. S. Mason Physiology and Bosch Institute for Medical Research, School of Medical Sciences, University of Sydney, Sydney, NSW 2006, Australia J. S. Chen : P. N. Sambrook : L. M. March : C. Kok : M. Macara Department of Rheumatology, Institute of Bone and Joint Research, University of Sydney, Royal North Shore Hospital, St Leonards, NSW 2065, Australia M. J. Seibel ANZAC Research Institute, Concord Hospital, University of Sydney, Sydney, NSW 2139, Australia
197, Page 2 of 7
Arch Osteoporos (2014) 9:197
Introduction
living in aged care facilities [17, 18]. However, artificial UV exposure is not practical on a large scale and others have attempted to model the equivalent sun exposure times. Holick [19] stated in his review that whole body exposure in a tanning bed is equivalent to sunlight exposure that produces 1 MED. Exposure of 15 % of the body surface area (e.g. hands, arms and face) to one third MED in the middle of the day can produce 1000 IU vitamin D, but duration of sun exposure depends on the latitude, season and skin type [15]. Webb and Engelsen [20] modelled the sun exposure times required to achieve a vitamin D dose of 1000 IU at different latitudes, seasons and times of day for different skin types. This model did not take into account the requirements for older people. Rhodes and colleagues [21] simulated midday summer sun exposure in participants who received the equivalent cumulative dose of 23.4 SED over 6 weeks. The mean increase in 25OHD was 26.0 nmol/L, but the weekly incremental increases in 25OHD reduced over the course of the intervention. Few studies have assessed the effects of natural sun exposure in older people on vitamin D levels. Reid et al. [22] randomly allocated 15 residents of an aged care facility in Auckland, New Zealand, to no sun exposure, 15 or 30 minutes of daily supervised sun exposure for 4 weeks in spring. A significant increase in 25OHD was noted in the 30-minute exposure group. Sato et al. conducted three randomized controlled trials (RCTs) of sun exposure in older people in Japan. Two studies were in long-term hospitalized patients with either Alzheimer’s disease or hemiplegic stroke [23, 24]. The participants had a supervised daily sun exposure for 12 months together with calcium supplements. Median 25OHD levels increased from 24 and 17–18 nmol/L, respectively, to 52 nmol/L. The third RCT involved elderly outpatients with Parkinson’s disease [25]. Participants in the intervention group were required to have 15 minutes of daily outdoor sun exposure and to record this on a calendar. Their 25OHD levels are also reported to have increased over a 2-year period. The UVR dose received was not measured in any of these studies. The primary aim of the current study was to determine UVR exposure and the vitamin D response of participants in a previous cluster RCT of sun exposure and calcium supplementation in older people in residential aged care facilities in Sydney, Australia [26]. Secondary outcome measures in the RCT included the Geriatric Depression Scale [27] a quality of life measure, EuroQoL (http://www.euroqol.org), selfreported health and hand grip strength. Of the two intervention groups, one group received sun exposure only, and the second group had sun exposure and calcium supplements. The sunlight exposure sessions were supervised by sunlight officers, and participants were asked to expose their face, arms and hands to the sun for 30–40 minutes in the morning for 5 days a week, for 12 months. During midsummer, the sunlight sessions were run between 8.30 and 9.30 a.m. and during midwinter between 9.30 and 11 a.m. Participants sat in a sun-
Vitamin D deficiency, which is common in older people in residential aged care, has been identified as an important risk factor for falls and fractures in this population [1]. Reduced sun exposure, inadequate synthesis of vitamin D from sun exposure and dietary deficiency contribute to vitamin D deficiency in older people [2, 3]. Vitamin D supplements, usually with calcium, have been shown to reduce the risk of falls and fractures in this population [4–6]. However, factors such as limited adherence with medication [7] and potential adverse effects, especially in a population where polypharmacy is common [8], may reduce the potential benefits of vitamin D supplements. As an alternative, the effectiveness of ultraviolet radiation in correcting vitamin D deficiency has been investigated. The main source of vitamin D in humans is through the action of sunlight-derived ultraviolet B radiation on the skin, converting 7-dehydrocholesterol (7-DHC) to pre vitamin D, which in turn is converted to the active form. Diehl and Chiu [9] noted a number of environmental, personal and behavioural factors that influence that vitamin D production from UV exposure. UV radiation (UVR) reaching the ground depends on the latitude, season, time of day, altitude, cloud cover and other local factors. Increased skin pigmentation, skin covering by clothing, decreased 7-DHC in older people [2] and sunscreen use also limit the amount of UVR available, or its ability to induce vitamin D synthesis. Minimal erythemal dose (MED) is the dose of UVR that produces mild reddening of the skin 16–24 h after exposure. As MED changes with skin type, standard erythemal dose (SED) has been adopted as a standardized measure of UVR [10]. SED does not rely on erythema and is independent of skin type. One SED is equivalent to erythemal effective radiant exposure of 100 J/m2. In people with pale skin, Fitzpatrick type I [11], 1 SED is equivalent to 0.5 MED. It has been estimated that indoor workers in northern Europe receive an annual solar UV exposure of about 200 SED [12], while Vishvakarman et al. [13] estimated the annual occupational exposures of mail delivery workers and physical education teachers in Queensland, Australia, to be 120–440 SED. The main measure of vitamin D in the body is 25hydroxyvitamin D (25OHD), and there is a debate about its optimal level. The Institute of Medicine guidelines state that most people have vitamin D sufficiency at levels of at least 50 nmol/L of 25OHD [14]. An Australian position statement has defined vitamin D adequacy as being ≥50 nmol/L 25OHD at the end of winter, with end of summer levels needing to be 10–20 nmol/L higher, to allow for the normal decrease during winter [15]. Others have proposed higher levels for bone and muscle health and nonclassical actions [16]. Controlled, artificial UVR has been shown to increase 25OHD levels in older people with vitamin D deficiency
Arch Osteoporos (2014) 9:197
exposed, outside area of the facility with the sunlight officer, engaging in sedentary activities such as talking, singing or having morning tea. Incidental sun exposure was not measured. Adherence with sunlight sessions was generally low, and overall, the study showed no significant reduction in falls or increase in 25OHD. However, in the 17 % of participants who attended more than 50 % of the sunlight sessions, there was a significant reduction in falls and a small but significant increase in serum 25OHD. However, only the number of sessions attended and not the UVR dose received by the participants was assessed in that study. Therefore, it was not clear if the ineffectiveness of the intervention was due only to inadequate attendance or whether the participants who attended had inadequate UVR exposure. The main aims of this secondary analysis of data from the RCT were to (1) estimate the UVR exposure of the participants (instead of attendance rates) during the first 6 months of the RCT and (2) determine the relationship between UVR exposure (as distinct from number of sessions attended), and the change in 25OHD from baseline to 6 months. As vitamin D response may attenuate over time [21], and as preliminary analysis showed that adherence with the intervention declined markedly in the second 6 months of the study, the vitamin D response for only the first 6 months was determined. As sunlight may influence other outcomes such as mood, secondary outcome measures of Geriatric Depression Scale (GDS), EuroQoL (EQ5D), self-reported health and hand grip strength were also assessed in terms of response to UVR exposure.
Methods This study used the sunlight session attendance records and other data from the participants in the sunlight RCT. For the purpose of the current analysis (except for parathyroid hormone levels), the sunlight only and sunlight plus calcium groups were combined. Those participants in whom 25OHD levels were not available, or who commenced vitamin D supplements during the course of the study, were excluded. The sunlight officers maintained the records of attendance for each participant, noting the dates, times and length of the sunlight sessions. Using these data and measurements from a UVR monitoring site in Sydney, the UVR data (measured by SEDs) were obtained from the Australian Radiation Protection and Nuclear Safety Agency (ARPANSA). ARPANSA maintains a network of UV monitors/dataloggers at capital cities and some major regional centres around Australia [28]. The units use a Solar Light UV biometer that measures erythemally effective ultraviolet radiation. Data are downloaded every minute and displayed live on the ARPANSA website (http://www.arpansa.gov.au/uvindex/ index.cfm). The ambient solar UV Index data for Sydney were used to calculate how much UVR was incident during
Page 3 of 7, 197
each of the exposure sessions for the participants of this study. This calculation took into account the ambient UVR at the date and time of day of the sessions, as well as the duration of sun exposure. This incident UVR can be considered as the available ambient UVR and is a maximum that could be received by any of the participants. Previous studies have shown that people receive approximately 15 to 30 % of available ambient UVR depending upon their activities while outside [29–31]. Monthly and six monthly available UVR totals for each participant for the times they were outside were calculated. Demographic, clinical and biochemical measures were obtained from the sunlight RCT study database. Total daily intake of calcium was calculated by adding 300 mg for each serve of dairy products (as reported by the participants at baseline), 300 mg assumed from other dietary sources and 600 mg for those taking calcium supplements. Descriptive statistics were used to summarise the demographic and clinical variables of participants at baseline and at 6 months and the total UVR exposure (measured by SED) over 6 months. Serum 25OHD was determined for each quartile of UVR dose. As 25OHD varies with age, sex and season of testing, a regression model of baseline 25OHD on age, sex and season was developed and used to calculate the residuals at baseline and 6 months. These residuals were then used in a multivariable linear regression model to assess the effects of SED on 25OHD at 6 months. The analysis took into account the cluster randomisation design using robust estimator and controlled for BMI, baseline 25OHD and calcium intake. In this analysis, variables of 25OHD and UVR exposure (SED) were logtransformed to satisfy the assumptions of the regression models. Therefore, increasing log SED (LnSED) by 1 would be associated with a proportionate increase of exponentiation of the coefficient in serum 25OHD. As there was progressive recruitment of aged care facilities, participants started their sunlight sessions in different seasons. It was hypothesized that the season of exposure would differentially affect the vitamin D response to UVR. Therefore, the interaction between LnSED and season of commencing sunlight sessions was tested and would be included in the final model if it was statistically significant. As 97.2 % of the participants had Fitzpatrick skin type I or II [11] (see Table 1), this variable was not included in the analysis. The sunlight RCT had the approval of the Research Ethics Committee of the Northern Sydney Central Coast Area Health Service.
Results There were 397 participants (190 in the sunlight only group and 207 in the sunlight and calcium group) across 34 residential aged care facilities. Of these participants, 105 people in whom 25OHD levels were not available at 6 months
197, Page 4 of 7
Arch Osteoporos (2014) 9:197
(e.g. due to death, transfer to a high care nursing home facility or loss to follow-up), 27 people who started vitamin D supplements during the first 6 months of starting the sunlight sessions and 13 people who attended no sessions during the 6 months were excluded. Four participants from one hostel had very high SED outlier values and the validity of their sun exposure times was in doubt. These four were also excluded from further analyses. Therefore, the final number of participants was 248. Baseline characteristics of participants are shown in Table 1. Table 2 shows clinical and biochemical variables at baseline and at 6 months. The median serum PTH decreased significantly in both the sunlight only and the sunlight plus calcium supplement groups from baseline to 6 months, while there was a small nonsignificant increase of 2.2 nmol/L in the median 25OHD. No significant changes from baseline to 6 months were observed in the GDS, EQ5D or in the
Table 1 Baseline characteristics of the intervention groups in the sunlight RCT Variable
Number Mean (±SD)/ number (%)
Age (years) Gender (female) BMI (kg/m2) Smoking Current smoker Ex-smoker Nonsmoker Alcohol use (≥12 drinks in the last 12 months) Dairy products/day 0 1
248 248 244 248
248
86.4 (±6.6) 179 (72.2 %) 26.4 (±5.2) 9 (3.6 %) 92 (37.1 %) 147 (59.3 %) 79 (31.9 %)
243
2 3 Daily calcium intake (diet and 243 supplements) 1000 mg Fitzpatrick skin type score (assessed at 245 12 months) Type I Type II Type III, IV Standardised Mini-Mental State Exami- 242 nation 0–17 18–23 24–30
12 (4.9 %) 95 (39.1 %) 95 (39.1 %) 41 (16.9 %)
72 (29.6 %) 70 (28.8 %) 101 (41.6 %)
143 (58.4 %) 95 (38.8 %) 7 (2.9 %)
22 (9.1 %) 64 (26.4 %) 156 (64.5 %)
proportion of those reporting excellent/very good/good health. There was a small significant increase in hand grip strength from baseline to 6 months. As seen in Table 3, the median SEDs for the participants who started their sunlight sessions in spring or autumn were 44.1 and 58.2, respectively. The median 25OHD levels at 6 months for those staring in spring and autumn were 34.5 and 33.5 nmol/L, respectively. The median 25OHD levels for each quartile of SED distribution are shown in Table 4. This was a nonsignificant trend (Kruskal-Wallis test p=0.054). Results of multivariable linear regression are shown in Table 5. The model accounted for 44 % of the variance in 25OHD at 6 months (F=16.6, p=0.000). Baseline 25OHD, which accounted for 33 % of the variance alone, was the largest predictor of 6-month 25OHD (adjusted coefficient= 0.594, 95 % confidence interval (CI) 0.465–0.724, p=0.00). The effect of UVR on 25OHD at 6 months depended on the season of exposure. The UVR dose for those who started sunlight sessions in spring or autumn had small but statistically significant positive effects on the 25OHD levels at 6 months. The coefficient of 0.105 for those who started sunlight sessions in spring indicates that an increase of 1 unit in LnSED was associated with an 11 % (e0.105=1.11) increase in 25OHD. The UVR dose for those who commenced sunlight sessions in summer or winter had nonsignificant negative effects on 25OHD levels at 6 months. Neither calcium intake nor BMI was a predictor of the 25OHD response.
Discussion This study estimated the ambient available UVR dose from therapeutic sunlight exposure of participants in the sunlight RCT by Sambrook et al. [26] and its effect on 25OHD levels in older people in residential aged care facilities. After 6 months of supervised sunlight exposure, there was a small nonsignificant increase in 25OHD levels. The season in which the participants had their sun exposure affected the 25OHD response to the UVR dose. For those who started their sunlight sessions in spring or autumn, UVR was a significant positive predictor of 25OHD at 6 months, while for those who started in summer or winter, UVR exposure had a nonsignificant negative effect on 25OHD. The study also determined that for those who started their sun exposure in spring, for every 1 unit increase in LnSED, there was an 11 % increase in 25OHD levels. The median SEDs for the spring and autumn groups were 44.1 and 58.2, respectively, and as noted by Gies et al. [29–31], participants could only expect to receive 15– 30 % of this available ambient UVR. Therefore, the actual UVR dose received would be lower, and certainly less than that reported by Rhodes et al. in their simulated sunlight
Arch Osteoporos (2014) 9:197 Table 2 Serum 25OHD, PTH and clinical characteristics of participants at baseline and 6 months
a
Wilcoxon rank-sum test
b
Paired t test
c
Chi-square test
Page 5 of 7, 197
Variable
Number
Serum 25-hydroxyvitamin D (nmol/L) Baseline 248 At 6 months assessment 248 Serum parathyroid hormone (pg/mL) Sunlight only group Baseline 107 At 6 months assessment 87 Sunlight and calcium group Baseline 122 At 6 months assessment 102 Geriatric Depression Scale Baseline 247 At 6 months assessment 248 EQ5D Baseline 248 At 6 months assessment 248 Hand grip strength (kg) Baseline 246 At 6 month assessment 243 Self-rated health (excellent/very good/good) Baseline 246 At 6 months assessment 248
exposure study [21], or in natural UV exposure in the general population [12, 13]. Despite the significant effect of season of exposure to 25OHD response, even at the end of summer, 25OHD levels were still below 50 nmol/L, the recommended level for vitamin D sufficiency, and well below the 60–70 nmol/L recommended at the end of summer to compensate for a winter decline [15]. This suboptimal response is most likely due to the low cumulative SED, which reflects generally poor adherence to the sunlight intervention. In the original RCT, the median adherence rate was 26 % of available sessions [26]. Several studies have shown increased 25OHD in response to short-term artificial UV radiation [17, 21, 32]. Many of these were conducted in people who were much younger than the participants in this study, and the response to UVR in older people may be attenuated [2, 9]. Further, in these studies,
Median (interquartile range)/mean (±SD)/number (%)
p valuea
32.4 (22.9–50.6) 34.6 (23.8–48.4)
0.35a
59.7 (44.4–92.1) 47.2 (37.0–70.9)
0.00a
62.2 (45.9–88.2) 54.9 (40.3–79.1)
0.003a
4.1(±3.1) 4.0 (±3.0)
0.82b
0.67 (±0.22) 0.69 (±0.23)
0.16b
19.9 (±8.0) 21.5 (±7.9)
0.00b
195 (79.3 %) 184 (74.2 %)
0.18c
known doses of UVR were delivered, in contrast to the current study, where the UVR dose received from sun exposure was subject to a number of modifying variables, including the degree of skin exposure. Others have reported increases in 25OHD levels after sun exposure but did not measure the UVR received by the participants [22–25]. The study by Reid et al. [22], which had very few participants, was of short duration and the generalisability of its results is uncertain. In one of the trials reported by Sato et al., the sun exposure was self-reported and did not appear to have been validated [25]. None of the three trials by Sato et al. [23–25] reported the season in which the vitamin D levels were assessed and nor did they state whether any of the participants started vitamin D supplements during the study period. A limitation to the current study was in the measurement of UVR exposure. The available ambient UVR was measured,
Table 3 Standard erythemal dose by season of starting sun exposure Variable
Number
Standard erythemal dose over 6 months Starting at spring 91 Starting at summer 48 Starting at autumn 41 Starting at winter 68 Total 248
Median (interquartile range)
44.1 (15.5–102.2) 55.8 (26.1–116.1) 58.2 (23.7–79.0) 72.5 (27.1–116.2) 57.5 (19.9–102)
Table 4 Median 25OHD levels by standard erythemal dose quartiles Standard erythemal dose (SED)
Median 25OHD (nmol/L)
First quartile (lowest) Second quartile Third quartile Fourth quartile (highest)
30.1 32.3 39.6 35.1
Kruskal-Wallis test p=0.054
197, Page 6 of 7
Arch Osteoporos (2014) 9:197
Table 5 Multivariate linear regression of vitamin D response to UV exposure Independent variables
Coefficient (95 % CI)
p value
LnD_baseline LnSED 6 months (spring) LnSED 6 months (summer) LnSED 6 months (autumn)
0.594 (0.465–0.724) 0.105 (0.001 to 0.209) −0.039 (−0.095 to 0.017) 0.056 (0.005 to 0.107)
0.00 0.05 0.16 0.03
LnSED 6 months (winter)
−0.040 (−0.126 to 0.045)
0.35
LnD log(vitamin D), LnSED log(standard erythemal dose)
but the dose reaching the participants depended on environmental conditions and on the degree of skin exposure, as noted by Diel and Chiu [9]. Similarly, Pettifor et al. noted higher 25OHD levels in elderly South African women in the months where hours of sunlight were fewer, but temperatures higher than at other times [33]. They concluded that this was due to more time spent outdoors with less skin covering by clothes during these warmer months. In the current study, it is assumed that most participants exposed their arms, hands and face. While the sessions were supervised by sunlight officers who instructed participants about suitable sun exposure, the adherence to skin exposure was not recorded. It is likely that those who had attended the sunlight sessions during the temperate seasons in Sydney were more willing to uncover their skin surface, without the discomfort of extreme heat or cold of the peak summer or winter sessions. Another possible factor in the poor 25OHD response is deficiency of 25OHD substrate. While older people have lower levels of 7-dehydrocholesterol in skin [2], Davie and Lawson showed that they have a similar capacity to younger people to produce vitamin D from normal UVR exposure [34]. As Nowson et al. [15] noted, the main reason for vitamin D deficiency in older people is due to limited sun exposure, rather than a reduced capacity to produce vitamin D from UVR. Despite the limitations, this study showed that older people have the capacity to produce vitamin D from natural sun exposure. The vitamin D response depended on the season, at least partly by influencing the degree of skin exposure. Although 25OHD levels did not reach sufficiency levels, the significant drop in PTH noted in the sunlight group may have other advantages, since higher PTH levels, even in the normal range, are associated with an increased risk of mortality [35]. While natural sunlight exposure for vitamin D deficiency does not appear to be a practical public health intervention in older people in residential aged care facilities, where there are attempts to encourage this, they should focus the timing for the start of this activity on the milder months of spring or autumn, as this timing was associated with a vitamin D response. The intervention should also target those people who are likely to have the lowest 25OHD concentrations. The percentage increase in serum vitamin D for every unit
of LnSED was able to be determined in this study and, together with accurate measurement of individual UVR exposure with UV tags, can provide practical guidance about levels of sun exposure required in different seasons for vitamin D production in older people. Practical and behavioural strategies may also improve adherence to sunlight exposure, and the outcomes of this study can inform policies about the use of outdoor spaces in residential aged care facilities.
Conflict of interest None. Funding The sunlight trial was supported by a project grant from the Australian National Health and Medical Research Council.
References 1. Flicker L, Mead K, MacInnis RJ, Nowson C, Scherer S, Stein MS, Thomasx J, Hopper JL, Wark JD (2003) Serum vitamin D and falls in older women in residential care in Australia. J Am Geriatr Soc 51: 1533–1538 2. MacLaughlin J, Holick MF (1985) Aging decreases the capacity of human skin to produce vitamin D3. J Clin Investig 76:1536–1538 3. Zochling J, Chen JS, Seibel M, Schwarz J, Cameron ID, Cumming RG, March L, Sambrook PN (2005) Calcium metabolism in the frail elderly. Clin Rheumatol 24:576–582 4. Bischoff-Ferrari HA, Willett WC, Wong JB, Stuck AE, Staehelin HB, Orav EJ, Thoma A, Kiel DP, Henschkowski J (2009) Prevention of nonvertebral fractures with oral vitamin D and dose dependency: a meta-analysis of randomized controlled trials. Arch Intern Med 169: 551–561 5. Cameron ID, Murray GR, Gillespie LD, Robertson MC, Hill KD, Cumming RG, Kerse N (2010) Interventions for preventing falls in older people in nursing care facilities and hospitals. Cochrane Database of Systematic Reviews CD005465 6. Chung M, Lee J, Terasawa T, Lau J, Trikalinos TA (2011) 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 155:827–838 7. Tsai KT, Chen JH, Wen CJ, Kuo HK, Lu IS, Chiu LS, Wu SC, Chan DC (2012) Medication adherence among geriatric outpatients prescribed multiple medications. Am J Geriatr Pharmacother 10:61–68 8. Slabaugh SL, Maio V, Templin M, Abouzaid S (2010) Prevalence and risk of polypharmacy among the elderly in an outpatient setting: a retrospective cohort study in the Emilia-Romagna region, Italy. Drugs Aging 27:1019–1028 9. Diehl JW, Chiu MW (2010) Effects of ambient sunlight and photoprotection on vitamin D status. Dermatol Ther 23:48–60 10. L’Eclairage CId (1999) Erythema reference action spectrum and standard erythema dose. CIE S0007/E-1998 11. Fitzpatrick TB (1988) The validity and practicality of sun-reactive skin types I through VI. Arch Dermatol 124:869–871 12. Diffey BL (2002) Human exposure to solar ultraviolet radiation. J Cosmet Dermatol 1:124–130 13. Vishvakarman D, Wong JC, Boreham BW (2001) Annual occupational exposure to ultraviolet radiation in central Queensland. Health Phys 81:536–544 14. Ross AC, Taylor CL, Yaktine AL, Valle HBD, Editors. Dietary reference intakes for calcium and vitamin D. Institute of Medicine. Washington DC: The National Academies Press; 2011. Available
Arch Osteoporos (2014) 9:197
15.
16. 17.
18.
19. 20. 21.
22.
23.
24.
from http://www.nap.edu/catalog.php?record_id=13050. Accessed 29 Jan 2014 Nowson CA, McGrath JJ, Ebeling PR et al (2012) Vitamin D and health in adults in Australia and New Zealand: a position statement. Med J Aust 196:686–687 Heaney RP, Holick MF (2011) Why the IOM recommendations for vitamin D are deficient. J Bone Miner Res 26:455–457 Chel VG, Ooms ME, Pavel S, de Gruijl F, Brand A, Lips P (2011) Prevention and treatment of vitamin D deficiency in Dutch psychogeriatric nursing home residents by weekly half-body UVB exposure after showering: a pilot study. Age Ageing 40:211–214 Chel VG, Ooms ME, Popp-Snijders C, Pavel S, Schothorst AA, Meulemans CC, Lips P (1998) Ultraviolet irradiation corrects vitamin D deficiency and suppresses secondary hyperparathyroidism in the elderly. J Bone Miner Res 13:1238–1242 Holick MF (2011) Vitamin D: a d-lightful solution for health. J Investig Med 59:872–880 Webb AR, Engelsen O (2006) Calculated ultraviolet exposure levels for a healthy vitamin D status. Photochem Photobiol 82:1697–1703 Rhodes LE, Webb AR, Fraser HI, Kift R, Durkin MT, Allan D, O’Brien SJ, Vail A, Berry JL (2010) Recommended summer sunlight exposure levels can produce sufficient (>or=20 ng ml(−1)) but not the proposed optimal (>or=32 ng ml(−1)) 25(OH)D levels at UK latitudes. J Investig Dermatol 130:1411–1418 Reid IR, Gallagher DJ, Bosworth J (1986) Prophylaxis against vitamin D deficiency in the elderly by regular sunlight exposure. Age Ageing 15:35–40 Sato Y, Iwamoto J, Kanoko T, Satoh K (2005) Amelioration of osteoporosis and hypovitaminosis D by sunlight exposure in hospitalized, elderly women with Alzheimer’s disease: a randomized controlled trial. J Bone Miner Res 20:1327–1333 Sato Y, Metoki N, Iwamoto J, Satoh K (2003) Amelioration of osteoporosis and hypovitaminosis D by sunlight exposure in stroke patients. Neurology 61:338–342, Erratum appears in Neurology. 2008 Mar 18;70(12):982
Page 7 of 7, 197 25. Sato Y, Iwamoto J, Honda Y (2011) Amelioration of osteoporosis and hypovitaminosis D by sunlight exposure in Parkinson’s disease. Parkinsonism Relat Disord 17:22–26 26. Sambrook PN, Cameron ID, Chen JS et al (2012) Does increased sunlight exposure work as a strategy to improve vitamin D status in the elderly: a cluster randomised controlled trial. Osteoporos Int 23: 615–624 27. Yesavage JA, Brink TL, Rose TL, Lum O, Huang V, Adey M, Leirer VO (1982) Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res 17:37–49 28. Gies P, Roy C, Javorniczky J, Henderson S, Lemus-Deschamps L, Driscoll C (2004) Global Solar UV Index: Australian measurements, forecasts and comparison with the UK. Photochem Photobiol 79:32–39 29. Gies P, Roy C, Toomey S, MacLennan R, Watson M (1998) Solar UVR exposures of primary school children at three locations in Queensland. Photochem Photobiol 68:78–83 30. Gies P, Wright J (2003) Measured solar ultraviolet radiation exposures of outdoor workers in Queensland in the building and construction industry. Photochem Photobiol 78:342–348 31. Gies P, Glanz K, O’Riordan D, Elliott T, Nehl E (2009) Measured occupational solar UVR exposures of lifeguards in pool settings. Am J Ind Med 52:645–653 32. Armas LA, Dowell S, Akhter M, Duthuluru S, Huerter C, Hollis BW, Lund R, Heaney RP (2007) Ultraviolet-B radiation increases serum 25-hydroxyvitamin D levels: the effect of UVB dose and skin color. J Am Acad Dermatol 57:588–593 33. Pettifor JM, Ross FP, Solomon L (1978) Seasonal variation in serum 25-hydroxycholecalciferol concentrations in elderly South African patients with fractures of femoral neck. Br Med J 1:826–827 34. Davie M, Lawson DE (1980) Assessment of plasma 25hydroxyvitamin D response to ultraviolet irradiation over a controlled area in young and elderly subjects. Clin Sci 58:235–242 35. Hagstrom E, Hellman P, Larsson TE et al (2009) Plasma parathyroid hormone and the risk of cardiovascular mortality in the community. Circulation 119:2765–2771
ORIGINAL ARTICLE
Vitamin D response of older people in residential aged care to sunlight-derived ultraviolet radiation Seeta Durvasula & Peter Gies & Rebecca S. Mason & Jian Sheng Chen & Stuart Henderson & Markus J. Seibel & Philip N. Sambrook & Lynette M. March & Stephen R. Lord & Cindy Kok & Monique Macara & Trevor R. Parmenter & Ian D. Cameron
Received: 28 February 2014 / Accepted: 24 September 2014 / Published online: 14 October 2014 # International Osteoporosis Foundation and National Osteoporosis Foundation 2014
S. R. Lord Neuroscience Research Australia, University of NSW, Randwick, NSW 2031, Australia
Purpose The purpose of this study is to measure the ultraviolet radiation (UVR) exposure of those in residential aged care in an earlier trial of sunlight exposure and to determine its effect on their vitamin D response. Methods Attendance data, demographic, clinical and biochemical variables for 248 participants were used for a secondary analysis of a previous cluster randomized trial of sunlight exposure and falls. The ambient solar UV Index data were used to calculate the participants’ UVR dose. Multiple linear regression was used to test if UVR exposure over 6 months, as measured by the standard erythemal dose (SED), was a predictor of vitamin D response, controlling for age, gender, BMI, calcium intake, baseline vitamin D and season of exposure. Results The median 25-hydroxyvitamin D (25OHD) was 32.4 nmol/L at baseline and 34.6 nmol/L at 6 months (p=0.35). The significant predictors of 25OHD at 6 months were UVR exposures in spring-summer (coefficient=0.105, 95 % confidence interval (CI) 0.001–0.209, p=0.05) and autumn-winter (coefficient=0.056, 95 % CI 0.005–0.107, p = 0.03) and baseline vitamin D (adjusted coefficient= 0.594, 95 % CI 0.465–0.724, p=0.00). In those starting sunlight sessions in spring, an increase of 1 unit in log SED was associated with 11 % increase in 25OHD. Conclusions Natural UVR exposure can increase 25OHD levels in older people in residential care, but depends on the season of exposure. However, due to inadequate sun exposure, 25OHD did not reach optimal levels. Nevertheless, where sun exposure is encouraged in this group, the focus for the start of exposure should be in the months of spring or autumn, as this timing was associated with a vitamin D response.
I. D. Cameron The John Walsh Centre for Rehabilitation Research, Sydney Medical School Northern, University of Sydney, St Leonards, NSW 2065, Australia
Keywords Sunlight . Ultraviolet radiation . Older people . Residential aged care . Vitamin D
Abstract Summary The aim of this study was to determine the vitamin D response to sunlight ultraviolet radiation in older people. Increases in vitamin D depended on the season of exposure, but the changes were small. Natural sun exposure is not a practical intervention for vitamin D deficiency in this population.
Philip N. Sambrook passed away during the course of this study. S. Durvasula (*) : T. R. Parmenter Sydney Medical School Northern, University of Sydney, Sydney, NSW 2006, Australia e-mail: [email protected] P. Gies : S. Henderson Australian Radiation Protection and Nuclear Safety Agency, 619 Lower Plenty Road, Yallambie, VIC 3085, Australia R. S. Mason Physiology and Bosch Institute for Medical Research, School of Medical Sciences, University of Sydney, Sydney, NSW 2006, Australia J. S. Chen : P. N. Sambrook : L. M. March : C. Kok : M. Macara Department of Rheumatology, Institute of Bone and Joint Research, University of Sydney, Royal North Shore Hospital, St Leonards, NSW 2065, Australia M. J. Seibel ANZAC Research Institute, Concord Hospital, University of Sydney, Sydney, NSW 2139, Australia
197, Page 2 of 7
Arch Osteoporos (2014) 9:197
Introduction
living in aged care facilities [17, 18]. However, artificial UV exposure is not practical on a large scale and others have attempted to model the equivalent sun exposure times. Holick [19] stated in his review that whole body exposure in a tanning bed is equivalent to sunlight exposure that produces 1 MED. Exposure of 15 % of the body surface area (e.g. hands, arms and face) to one third MED in the middle of the day can produce 1000 IU vitamin D, but duration of sun exposure depends on the latitude, season and skin type [15]. Webb and Engelsen [20] modelled the sun exposure times required to achieve a vitamin D dose of 1000 IU at different latitudes, seasons and times of day for different skin types. This model did not take into account the requirements for older people. Rhodes and colleagues [21] simulated midday summer sun exposure in participants who received the equivalent cumulative dose of 23.4 SED over 6 weeks. The mean increase in 25OHD was 26.0 nmol/L, but the weekly incremental increases in 25OHD reduced over the course of the intervention. Few studies have assessed the effects of natural sun exposure in older people on vitamin D levels. Reid et al. [22] randomly allocated 15 residents of an aged care facility in Auckland, New Zealand, to no sun exposure, 15 or 30 minutes of daily supervised sun exposure for 4 weeks in spring. A significant increase in 25OHD was noted in the 30-minute exposure group. Sato et al. conducted three randomized controlled trials (RCTs) of sun exposure in older people in Japan. Two studies were in long-term hospitalized patients with either Alzheimer’s disease or hemiplegic stroke [23, 24]. The participants had a supervised daily sun exposure for 12 months together with calcium supplements. Median 25OHD levels increased from 24 and 17–18 nmol/L, respectively, to 52 nmol/L. The third RCT involved elderly outpatients with Parkinson’s disease [25]. Participants in the intervention group were required to have 15 minutes of daily outdoor sun exposure and to record this on a calendar. Their 25OHD levels are also reported to have increased over a 2-year period. The UVR dose received was not measured in any of these studies. The primary aim of the current study was to determine UVR exposure and the vitamin D response of participants in a previous cluster RCT of sun exposure and calcium supplementation in older people in residential aged care facilities in Sydney, Australia [26]. Secondary outcome measures in the RCT included the Geriatric Depression Scale [27] a quality of life measure, EuroQoL (http://www.euroqol.org), selfreported health and hand grip strength. Of the two intervention groups, one group received sun exposure only, and the second group had sun exposure and calcium supplements. The sunlight exposure sessions were supervised by sunlight officers, and participants were asked to expose their face, arms and hands to the sun for 30–40 minutes in the morning for 5 days a week, for 12 months. During midsummer, the sunlight sessions were run between 8.30 and 9.30 a.m. and during midwinter between 9.30 and 11 a.m. Participants sat in a sun-
Vitamin D deficiency, which is common in older people in residential aged care, has been identified as an important risk factor for falls and fractures in this population [1]. Reduced sun exposure, inadequate synthesis of vitamin D from sun exposure and dietary deficiency contribute to vitamin D deficiency in older people [2, 3]. Vitamin D supplements, usually with calcium, have been shown to reduce the risk of falls and fractures in this population [4–6]. However, factors such as limited adherence with medication [7] and potential adverse effects, especially in a population where polypharmacy is common [8], may reduce the potential benefits of vitamin D supplements. As an alternative, the effectiveness of ultraviolet radiation in correcting vitamin D deficiency has been investigated. The main source of vitamin D in humans is through the action of sunlight-derived ultraviolet B radiation on the skin, converting 7-dehydrocholesterol (7-DHC) to pre vitamin D, which in turn is converted to the active form. Diehl and Chiu [9] noted a number of environmental, personal and behavioural factors that influence that vitamin D production from UV exposure. UV radiation (UVR) reaching the ground depends on the latitude, season, time of day, altitude, cloud cover and other local factors. Increased skin pigmentation, skin covering by clothing, decreased 7-DHC in older people [2] and sunscreen use also limit the amount of UVR available, or its ability to induce vitamin D synthesis. Minimal erythemal dose (MED) is the dose of UVR that produces mild reddening of the skin 16–24 h after exposure. As MED changes with skin type, standard erythemal dose (SED) has been adopted as a standardized measure of UVR [10]. SED does not rely on erythema and is independent of skin type. One SED is equivalent to erythemal effective radiant exposure of 100 J/m2. In people with pale skin, Fitzpatrick type I [11], 1 SED is equivalent to 0.5 MED. It has been estimated that indoor workers in northern Europe receive an annual solar UV exposure of about 200 SED [12], while Vishvakarman et al. [13] estimated the annual occupational exposures of mail delivery workers and physical education teachers in Queensland, Australia, to be 120–440 SED. The main measure of vitamin D in the body is 25hydroxyvitamin D (25OHD), and there is a debate about its optimal level. The Institute of Medicine guidelines state that most people have vitamin D sufficiency at levels of at least 50 nmol/L of 25OHD [14]. An Australian position statement has defined vitamin D adequacy as being ≥50 nmol/L 25OHD at the end of winter, with end of summer levels needing to be 10–20 nmol/L higher, to allow for the normal decrease during winter [15]. Others have proposed higher levels for bone and muscle health and nonclassical actions [16]. Controlled, artificial UVR has been shown to increase 25OHD levels in older people with vitamin D deficiency
Arch Osteoporos (2014) 9:197
exposed, outside area of the facility with the sunlight officer, engaging in sedentary activities such as talking, singing or having morning tea. Incidental sun exposure was not measured. Adherence with sunlight sessions was generally low, and overall, the study showed no significant reduction in falls or increase in 25OHD. However, in the 17 % of participants who attended more than 50 % of the sunlight sessions, there was a significant reduction in falls and a small but significant increase in serum 25OHD. However, only the number of sessions attended and not the UVR dose received by the participants was assessed in that study. Therefore, it was not clear if the ineffectiveness of the intervention was due only to inadequate attendance or whether the participants who attended had inadequate UVR exposure. The main aims of this secondary analysis of data from the RCT were to (1) estimate the UVR exposure of the participants (instead of attendance rates) during the first 6 months of the RCT and (2) determine the relationship between UVR exposure (as distinct from number of sessions attended), and the change in 25OHD from baseline to 6 months. As vitamin D response may attenuate over time [21], and as preliminary analysis showed that adherence with the intervention declined markedly in the second 6 months of the study, the vitamin D response for only the first 6 months was determined. As sunlight may influence other outcomes such as mood, secondary outcome measures of Geriatric Depression Scale (GDS), EuroQoL (EQ5D), self-reported health and hand grip strength were also assessed in terms of response to UVR exposure.
Methods This study used the sunlight session attendance records and other data from the participants in the sunlight RCT. For the purpose of the current analysis (except for parathyroid hormone levels), the sunlight only and sunlight plus calcium groups were combined. Those participants in whom 25OHD levels were not available, or who commenced vitamin D supplements during the course of the study, were excluded. The sunlight officers maintained the records of attendance for each participant, noting the dates, times and length of the sunlight sessions. Using these data and measurements from a UVR monitoring site in Sydney, the UVR data (measured by SEDs) were obtained from the Australian Radiation Protection and Nuclear Safety Agency (ARPANSA). ARPANSA maintains a network of UV monitors/dataloggers at capital cities and some major regional centres around Australia [28]. The units use a Solar Light UV biometer that measures erythemally effective ultraviolet radiation. Data are downloaded every minute and displayed live on the ARPANSA website (http://www.arpansa.gov.au/uvindex/ index.cfm). The ambient solar UV Index data for Sydney were used to calculate how much UVR was incident during
Page 3 of 7, 197
each of the exposure sessions for the participants of this study. This calculation took into account the ambient UVR at the date and time of day of the sessions, as well as the duration of sun exposure. This incident UVR can be considered as the available ambient UVR and is a maximum that could be received by any of the participants. Previous studies have shown that people receive approximately 15 to 30 % of available ambient UVR depending upon their activities while outside [29–31]. Monthly and six monthly available UVR totals for each participant for the times they were outside were calculated. Demographic, clinical and biochemical measures were obtained from the sunlight RCT study database. Total daily intake of calcium was calculated by adding 300 mg for each serve of dairy products (as reported by the participants at baseline), 300 mg assumed from other dietary sources and 600 mg for those taking calcium supplements. Descriptive statistics were used to summarise the demographic and clinical variables of participants at baseline and at 6 months and the total UVR exposure (measured by SED) over 6 months. Serum 25OHD was determined for each quartile of UVR dose. As 25OHD varies with age, sex and season of testing, a regression model of baseline 25OHD on age, sex and season was developed and used to calculate the residuals at baseline and 6 months. These residuals were then used in a multivariable linear regression model to assess the effects of SED on 25OHD at 6 months. The analysis took into account the cluster randomisation design using robust estimator and controlled for BMI, baseline 25OHD and calcium intake. In this analysis, variables of 25OHD and UVR exposure (SED) were logtransformed to satisfy the assumptions of the regression models. Therefore, increasing log SED (LnSED) by 1 would be associated with a proportionate increase of exponentiation of the coefficient in serum 25OHD. As there was progressive recruitment of aged care facilities, participants started their sunlight sessions in different seasons. It was hypothesized that the season of exposure would differentially affect the vitamin D response to UVR. Therefore, the interaction between LnSED and season of commencing sunlight sessions was tested and would be included in the final model if it was statistically significant. As 97.2 % of the participants had Fitzpatrick skin type I or II [11] (see Table 1), this variable was not included in the analysis. The sunlight RCT had the approval of the Research Ethics Committee of the Northern Sydney Central Coast Area Health Service.
Results There were 397 participants (190 in the sunlight only group and 207 in the sunlight and calcium group) across 34 residential aged care facilities. Of these participants, 105 people in whom 25OHD levels were not available at 6 months
197, Page 4 of 7
Arch Osteoporos (2014) 9:197
(e.g. due to death, transfer to a high care nursing home facility or loss to follow-up), 27 people who started vitamin D supplements during the first 6 months of starting the sunlight sessions and 13 people who attended no sessions during the 6 months were excluded. Four participants from one hostel had very high SED outlier values and the validity of their sun exposure times was in doubt. These four were also excluded from further analyses. Therefore, the final number of participants was 248. Baseline characteristics of participants are shown in Table 1. Table 2 shows clinical and biochemical variables at baseline and at 6 months. The median serum PTH decreased significantly in both the sunlight only and the sunlight plus calcium supplement groups from baseline to 6 months, while there was a small nonsignificant increase of 2.2 nmol/L in the median 25OHD. No significant changes from baseline to 6 months were observed in the GDS, EQ5D or in the
Table 1 Baseline characteristics of the intervention groups in the sunlight RCT Variable
Number Mean (±SD)/ number (%)
Age (years) Gender (female) BMI (kg/m2) Smoking Current smoker Ex-smoker Nonsmoker Alcohol use (≥12 drinks in the last 12 months) Dairy products/day 0 1
248 248 244 248
248
86.4 (±6.6) 179 (72.2 %) 26.4 (±5.2) 9 (3.6 %) 92 (37.1 %) 147 (59.3 %) 79 (31.9 %)
243
2 3 Daily calcium intake (diet and 243 supplements) 1000 mg Fitzpatrick skin type score (assessed at 245 12 months) Type I Type II Type III, IV Standardised Mini-Mental State Exami- 242 nation 0–17 18–23 24–30
12 (4.9 %) 95 (39.1 %) 95 (39.1 %) 41 (16.9 %)
72 (29.6 %) 70 (28.8 %) 101 (41.6 %)
143 (58.4 %) 95 (38.8 %) 7 (2.9 %)
22 (9.1 %) 64 (26.4 %) 156 (64.5 %)
proportion of those reporting excellent/very good/good health. There was a small significant increase in hand grip strength from baseline to 6 months. As seen in Table 3, the median SEDs for the participants who started their sunlight sessions in spring or autumn were 44.1 and 58.2, respectively. The median 25OHD levels at 6 months for those staring in spring and autumn were 34.5 and 33.5 nmol/L, respectively. The median 25OHD levels for each quartile of SED distribution are shown in Table 4. This was a nonsignificant trend (Kruskal-Wallis test p=0.054). Results of multivariable linear regression are shown in Table 5. The model accounted for 44 % of the variance in 25OHD at 6 months (F=16.6, p=0.000). Baseline 25OHD, which accounted for 33 % of the variance alone, was the largest predictor of 6-month 25OHD (adjusted coefficient= 0.594, 95 % confidence interval (CI) 0.465–0.724, p=0.00). The effect of UVR on 25OHD at 6 months depended on the season of exposure. The UVR dose for those who started sunlight sessions in spring or autumn had small but statistically significant positive effects on the 25OHD levels at 6 months. The coefficient of 0.105 for those who started sunlight sessions in spring indicates that an increase of 1 unit in LnSED was associated with an 11 % (e0.105=1.11) increase in 25OHD. The UVR dose for those who commenced sunlight sessions in summer or winter had nonsignificant negative effects on 25OHD levels at 6 months. Neither calcium intake nor BMI was a predictor of the 25OHD response.
Discussion This study estimated the ambient available UVR dose from therapeutic sunlight exposure of participants in the sunlight RCT by Sambrook et al. [26] and its effect on 25OHD levels in older people in residential aged care facilities. After 6 months of supervised sunlight exposure, there was a small nonsignificant increase in 25OHD levels. The season in which the participants had their sun exposure affected the 25OHD response to the UVR dose. For those who started their sunlight sessions in spring or autumn, UVR was a significant positive predictor of 25OHD at 6 months, while for those who started in summer or winter, UVR exposure had a nonsignificant negative effect on 25OHD. The study also determined that for those who started their sun exposure in spring, for every 1 unit increase in LnSED, there was an 11 % increase in 25OHD levels. The median SEDs for the spring and autumn groups were 44.1 and 58.2, respectively, and as noted by Gies et al. [29–31], participants could only expect to receive 15– 30 % of this available ambient UVR. Therefore, the actual UVR dose received would be lower, and certainly less than that reported by Rhodes et al. in their simulated sunlight
Arch Osteoporos (2014) 9:197 Table 2 Serum 25OHD, PTH and clinical characteristics of participants at baseline and 6 months
a
Wilcoxon rank-sum test
b
Paired t test
c
Chi-square test
Page 5 of 7, 197
Variable
Number
Serum 25-hydroxyvitamin D (nmol/L) Baseline 248 At 6 months assessment 248 Serum parathyroid hormone (pg/mL) Sunlight only group Baseline 107 At 6 months assessment 87 Sunlight and calcium group Baseline 122 At 6 months assessment 102 Geriatric Depression Scale Baseline 247 At 6 months assessment 248 EQ5D Baseline 248 At 6 months assessment 248 Hand grip strength (kg) Baseline 246 At 6 month assessment 243 Self-rated health (excellent/very good/good) Baseline 246 At 6 months assessment 248
exposure study [21], or in natural UV exposure in the general population [12, 13]. Despite the significant effect of season of exposure to 25OHD response, even at the end of summer, 25OHD levels were still below 50 nmol/L, the recommended level for vitamin D sufficiency, and well below the 60–70 nmol/L recommended at the end of summer to compensate for a winter decline [15]. This suboptimal response is most likely due to the low cumulative SED, which reflects generally poor adherence to the sunlight intervention. In the original RCT, the median adherence rate was 26 % of available sessions [26]. Several studies have shown increased 25OHD in response to short-term artificial UV radiation [17, 21, 32]. Many of these were conducted in people who were much younger than the participants in this study, and the response to UVR in older people may be attenuated [2, 9]. Further, in these studies,
Median (interquartile range)/mean (±SD)/number (%)
p valuea
32.4 (22.9–50.6) 34.6 (23.8–48.4)
0.35a
59.7 (44.4–92.1) 47.2 (37.0–70.9)
0.00a
62.2 (45.9–88.2) 54.9 (40.3–79.1)
0.003a
4.1(±3.1) 4.0 (±3.0)
0.82b
0.67 (±0.22) 0.69 (±0.23)
0.16b
19.9 (±8.0) 21.5 (±7.9)
0.00b
195 (79.3 %) 184 (74.2 %)
0.18c
known doses of UVR were delivered, in contrast to the current study, where the UVR dose received from sun exposure was subject to a number of modifying variables, including the degree of skin exposure. Others have reported increases in 25OHD levels after sun exposure but did not measure the UVR received by the participants [22–25]. The study by Reid et al. [22], which had very few participants, was of short duration and the generalisability of its results is uncertain. In one of the trials reported by Sato et al., the sun exposure was self-reported and did not appear to have been validated [25]. None of the three trials by Sato et al. [23–25] reported the season in which the vitamin D levels were assessed and nor did they state whether any of the participants started vitamin D supplements during the study period. A limitation to the current study was in the measurement of UVR exposure. The available ambient UVR was measured,
Table 3 Standard erythemal dose by season of starting sun exposure Variable
Number
Standard erythemal dose over 6 months Starting at spring 91 Starting at summer 48 Starting at autumn 41 Starting at winter 68 Total 248
Median (interquartile range)
44.1 (15.5–102.2) 55.8 (26.1–116.1) 58.2 (23.7–79.0) 72.5 (27.1–116.2) 57.5 (19.9–102)
Table 4 Median 25OHD levels by standard erythemal dose quartiles Standard erythemal dose (SED)
Median 25OHD (nmol/L)
First quartile (lowest) Second quartile Third quartile Fourth quartile (highest)
30.1 32.3 39.6 35.1
Kruskal-Wallis test p=0.054
197, Page 6 of 7
Arch Osteoporos (2014) 9:197
Table 5 Multivariate linear regression of vitamin D response to UV exposure Independent variables
Coefficient (95 % CI)
p value
LnD_baseline LnSED 6 months (spring) LnSED 6 months (summer) LnSED 6 months (autumn)
0.594 (0.465–0.724) 0.105 (0.001 to 0.209) −0.039 (−0.095 to 0.017) 0.056 (0.005 to 0.107)
0.00 0.05 0.16 0.03
LnSED 6 months (winter)
−0.040 (−0.126 to 0.045)
0.35
LnD log(vitamin D), LnSED log(standard erythemal dose)
but the dose reaching the participants depended on environmental conditions and on the degree of skin exposure, as noted by Diel and Chiu [9]. Similarly, Pettifor et al. noted higher 25OHD levels in elderly South African women in the months where hours of sunlight were fewer, but temperatures higher than at other times [33]. They concluded that this was due to more time spent outdoors with less skin covering by clothes during these warmer months. In the current study, it is assumed that most participants exposed their arms, hands and face. While the sessions were supervised by sunlight officers who instructed participants about suitable sun exposure, the adherence to skin exposure was not recorded. It is likely that those who had attended the sunlight sessions during the temperate seasons in Sydney were more willing to uncover their skin surface, without the discomfort of extreme heat or cold of the peak summer or winter sessions. Another possible factor in the poor 25OHD response is deficiency of 25OHD substrate. While older people have lower levels of 7-dehydrocholesterol in skin [2], Davie and Lawson showed that they have a similar capacity to younger people to produce vitamin D from normal UVR exposure [34]. As Nowson et al. [15] noted, the main reason for vitamin D deficiency in older people is due to limited sun exposure, rather than a reduced capacity to produce vitamin D from UVR. Despite the limitations, this study showed that older people have the capacity to produce vitamin D from natural sun exposure. The vitamin D response depended on the season, at least partly by influencing the degree of skin exposure. Although 25OHD levels did not reach sufficiency levels, the significant drop in PTH noted in the sunlight group may have other advantages, since higher PTH levels, even in the normal range, are associated with an increased risk of mortality [35]. While natural sunlight exposure for vitamin D deficiency does not appear to be a practical public health intervention in older people in residential aged care facilities, where there are attempts to encourage this, they should focus the timing for the start of this activity on the milder months of spring or autumn, as this timing was associated with a vitamin D response. The intervention should also target those people who are likely to have the lowest 25OHD concentrations. The percentage increase in serum vitamin D for every unit
of LnSED was able to be determined in this study and, together with accurate measurement of individual UVR exposure with UV tags, can provide practical guidance about levels of sun exposure required in different seasons for vitamin D production in older people. Practical and behavioural strategies may also improve adherence to sunlight exposure, and the outcomes of this study can inform policies about the use of outdoor spaces in residential aged care facilities.
Conflict of interest None. Funding The sunlight trial was supported by a project grant from the Australian National Health and Medical Research Council.
References 1. Flicker L, Mead K, MacInnis RJ, Nowson C, Scherer S, Stein MS, Thomasx J, Hopper JL, Wark JD (2003) Serum vitamin D and falls in older women in residential care in Australia. J Am Geriatr Soc 51: 1533–1538 2. MacLaughlin J, Holick MF (1985) Aging decreases the capacity of human skin to produce vitamin D3. J Clin Investig 76:1536–1538 3. Zochling J, Chen JS, Seibel M, Schwarz J, Cameron ID, Cumming RG, March L, Sambrook PN (2005) Calcium metabolism in the frail elderly. Clin Rheumatol 24:576–582 4. Bischoff-Ferrari HA, Willett WC, Wong JB, Stuck AE, Staehelin HB, Orav EJ, Thoma A, Kiel DP, Henschkowski J (2009) Prevention of nonvertebral fractures with oral vitamin D and dose dependency: a meta-analysis of randomized controlled trials. Arch Intern Med 169: 551–561 5. Cameron ID, Murray GR, Gillespie LD, Robertson MC, Hill KD, Cumming RG, Kerse N (2010) Interventions for preventing falls in older people in nursing care facilities and hospitals. Cochrane Database of Systematic Reviews CD005465 6. Chung M, Lee J, Terasawa T, Lau J, Trikalinos TA (2011) 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 155:827–838 7. Tsai KT, Chen JH, Wen CJ, Kuo HK, Lu IS, Chiu LS, Wu SC, Chan DC (2012) Medication adherence among geriatric outpatients prescribed multiple medications. Am J Geriatr Pharmacother 10:61–68 8. Slabaugh SL, Maio V, Templin M, Abouzaid S (2010) Prevalence and risk of polypharmacy among the elderly in an outpatient setting: a retrospective cohort study in the Emilia-Romagna region, Italy. Drugs Aging 27:1019–1028 9. Diehl JW, Chiu MW (2010) Effects of ambient sunlight and photoprotection on vitamin D status. Dermatol Ther 23:48–60 10. L’Eclairage CId (1999) Erythema reference action spectrum and standard erythema dose. CIE S0007/E-1998 11. Fitzpatrick TB (1988) The validity and practicality of sun-reactive skin types I through VI. Arch Dermatol 124:869–871 12. Diffey BL (2002) Human exposure to solar ultraviolet radiation. J Cosmet Dermatol 1:124–130 13. Vishvakarman D, Wong JC, Boreham BW (2001) Annual occupational exposure to ultraviolet radiation in central Queensland. Health Phys 81:536–544 14. Ross AC, Taylor CL, Yaktine AL, Valle HBD, Editors. Dietary reference intakes for calcium and vitamin D. Institute of Medicine. Washington DC: The National Academies Press; 2011. Available
Arch Osteoporos (2014) 9:197
15.
16. 17.
18.
19. 20. 21.
22.
23.
24.
from http://www.nap.edu/catalog.php?record_id=13050. Accessed 29 Jan 2014 Nowson CA, McGrath JJ, Ebeling PR et al (2012) Vitamin D and health in adults in Australia and New Zealand: a position statement. Med J Aust 196:686–687 Heaney RP, Holick MF (2011) Why the IOM recommendations for vitamin D are deficient. J Bone Miner Res 26:455–457 Chel VG, Ooms ME, Pavel S, de Gruijl F, Brand A, Lips P (2011) Prevention and treatment of vitamin D deficiency in Dutch psychogeriatric nursing home residents by weekly half-body UVB exposure after showering: a pilot study. Age Ageing 40:211–214 Chel VG, Ooms ME, Popp-Snijders C, Pavel S, Schothorst AA, Meulemans CC, Lips P (1998) Ultraviolet irradiation corrects vitamin D deficiency and suppresses secondary hyperparathyroidism in the elderly. J Bone Miner Res 13:1238–1242 Holick MF (2011) Vitamin D: a d-lightful solution for health. J Investig Med 59:872–880 Webb AR, Engelsen O (2006) Calculated ultraviolet exposure levels for a healthy vitamin D status. Photochem Photobiol 82:1697–1703 Rhodes LE, Webb AR, Fraser HI, Kift R, Durkin MT, Allan D, O’Brien SJ, Vail A, Berry JL (2010) Recommended summer sunlight exposure levels can produce sufficient (>or=20 ng ml(−1)) but not the proposed optimal (>or=32 ng ml(−1)) 25(OH)D levels at UK latitudes. J Investig Dermatol 130:1411–1418 Reid IR, Gallagher DJ, Bosworth J (1986) Prophylaxis against vitamin D deficiency in the elderly by regular sunlight exposure. Age Ageing 15:35–40 Sato Y, Iwamoto J, Kanoko T, Satoh K (2005) Amelioration of osteoporosis and hypovitaminosis D by sunlight exposure in hospitalized, elderly women with Alzheimer’s disease: a randomized controlled trial. J Bone Miner Res 20:1327–1333 Sato Y, Metoki N, Iwamoto J, Satoh K (2003) Amelioration of osteoporosis and hypovitaminosis D by sunlight exposure in stroke patients. Neurology 61:338–342, Erratum appears in Neurology. 2008 Mar 18;70(12):982
Page 7 of 7, 197 25. Sato Y, Iwamoto J, Honda Y (2011) Amelioration of osteoporosis and hypovitaminosis D by sunlight exposure in Parkinson’s disease. Parkinsonism Relat Disord 17:22–26 26. Sambrook PN, Cameron ID, Chen JS et al (2012) Does increased sunlight exposure work as a strategy to improve vitamin D status in the elderly: a cluster randomised controlled trial. Osteoporos Int 23: 615–624 27. Yesavage JA, Brink TL, Rose TL, Lum O, Huang V, Adey M, Leirer VO (1982) Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res 17:37–49 28. Gies P, Roy C, Javorniczky J, Henderson S, Lemus-Deschamps L, Driscoll C (2004) Global Solar UV Index: Australian measurements, forecasts and comparison with the UK. Photochem Photobiol 79:32–39 29. Gies P, Roy C, Toomey S, MacLennan R, Watson M (1998) Solar UVR exposures of primary school children at three locations in Queensland. Photochem Photobiol 68:78–83 30. Gies P, Wright J (2003) Measured solar ultraviolet radiation exposures of outdoor workers in Queensland in the building and construction industry. Photochem Photobiol 78:342–348 31. Gies P, Glanz K, O’Riordan D, Elliott T, Nehl E (2009) Measured occupational solar UVR exposures of lifeguards in pool settings. Am J Ind Med 52:645–653 32. Armas LA, Dowell S, Akhter M, Duthuluru S, Huerter C, Hollis BW, Lund R, Heaney RP (2007) Ultraviolet-B radiation increases serum 25-hydroxyvitamin D levels: the effect of UVB dose and skin color. J Am Acad Dermatol 57:588–593 33. Pettifor JM, Ross FP, Solomon L (1978) Seasonal variation in serum 25-hydroxycholecalciferol concentrations in elderly South African patients with fractures of femoral neck. Br Med J 1:826–827 34. Davie M, Lawson DE (1980) Assessment of plasma 25hydroxyvitamin D response to ultraviolet irradiation over a controlled area in young and elderly subjects. Clin Sci 58:235–242 35. Hagstrom E, Hellman P, Larsson TE et al (2009) Plasma parathyroid hormone and the risk of cardiovascular mortality in the community. Circulation 119:2765–2771
