Dietary Supplements

Development and Internal Validation of a Clinical Risk Score for Hypovitaminosis D in Italian Adults Aged ≥50 Years Attending Osteoporosis and Metabolic Bone Disease Centers.

TL;DR

A clinical risk score for predicting hypovitaminosis-D (25(OH)D < 20 ng/mL) based on eight easily assessable risk factors showed good discrimination (AUC: 79.1%) and calibration (r = 0.98) in Italian adults aged ≥50 years attending osteoporosis and metabolic bone disease centers.

Key Findings

The study population had a median 25(OH)D of 33.2 ng/mL, with 9.8% of subjects having levels below the primary deficiency threshold of 20 ng/mL.

  • The study included 1408 adults aged ≥50 years (1286 women and 122 men) attending centers across Italy dedicated to osteoporosis and metabolic bone diseases.
  • Median age was 67 years and 91.3% were female.
  • Overall, 1147 subjects (81.5%) were receiving cholecalciferol supplementation at the time of the study.
  • The cross-sectional design used a standardized questionnaire to collect demographic, clinical, lifestyle, and dietary data.

Eight independent predictors of hypovitaminosis-D (25(OH)D < 20 ng/mL) were identified through multivariable logistic regression.

  • Independent predictors included: higher body mass index, residence in Northern Italy, reduced summer sun exposure, sunscreen use, cardiovascular disease, glucocorticoid use, absence of cholecalciferol supplementation, and no prior vitamin D use.
  • Both univariable and multivariable logistic regression analyses were used to identify these predictors.
  • All predictors were based on easily assessable clinical and lifestyle factors collected via standardized questionnaire.
  • A second model was also developed for the secondary outcome of 25(OH)D < 30 ng/mL.

The derived clinical risk score (range 9–18) demonstrated good discriminative ability for predicting 25(OH)D < 20 ng/mL.

  • Area under the ROC curve (AUC) was 79.1% (95% CI: 75.3–82.9%).
  • Calibration was strong, with a correlation between predicted and observed probabilities of r = 0.98 (p < 0.001).
  • The score was internally validated.
  • Performance remained stable across seasons and in untreated subjects, with an AUC of 72.7% in the untreated subgroup.

Two distinct cut-off ranges of the risk score were identified to serve different clinical screening purposes.

  • A screening cut-off of 10.3–10.7 ensured high sensitivity of 87.0–92.7%, suitable for identifying individuals unlikely to be deficient.
  • A cut-off of 11.9–12.0 balanced sensitivity (~62%) and specificity (~80%), suited for targeted diagnostic testing.
  • The score range was 9 to 18 points.
  • These two cut-off strategies correspond to different clinical use cases: ruling out deficiency versus identifying high-risk individuals.

The risk score for the secondary outcome of 25(OH)D < 30 ng/mL showed only moderate discrimination.

  • The AUC for the secondary score predicting 25(OH)D < 30 ng/mL was 69.6%.
  • This is lower than the 79.1% AUC achieved for the primary outcome of 25(OH)D < 20 ng/mL.
  • The authors reported this as moderate discrimination compared to the good discrimination for the primary outcome.

The study identified several limitations that restrict the generalizability of the risk score.

  • Female predominance (91.3%) in the sample limits applicability to men.
  • Widespread vitamin D supplementation use (81.5% of subjects) may affect generalizability to populations with lower supplementation rates.
  • The absence of external validation in an independent cohort is a key limitation.
  • No impact analyses were conducted to assess whether use of the score changes clinical outcomes or testing behavior.
  • The study population was restricted to those attending specialized osteoporosis and metabolic bone disease centers in Italy.

The authors propose that the risk score may support targeted screening strategies and potentially reduce unnecessary vitamin D testing in routine clinical practice.

  • Increasing demand for serum 25(OH)D testing was cited as a motivating factor for developing simpler risk identification tools.
  • The score is based on easily assessable risk factors, intended for use without laboratory testing.
  • The authors state the score 'may help identify individuals at risk of vitamin D deficiency and support targeted screening strategies, potentially reducing unnecessary testing in routine clinical practice.'
  • Hypovitaminosis-D is described as 'a highly prevalent condition worldwide, associated with adverse skeletal and extra-skeletal outcomes.'

What This Means

This research suggests that a simple scoring tool based on eight easily observable characteristics — including body weight, where someone lives in Italy, how much time they spend in the sun in summer, sunscreen use, heart disease history, steroid medication use, and whether they take vitamin D supplements — can reasonably predict whether an adult aged 50 or older is likely to have low vitamin D levels (below 20 ng/mL). The score was developed and tested in a group of 1,408 Italian adults visiting clinics for bone health concerns, the majority of whom were women (91%) with a median age of 67. The tool correctly identified people at risk about 79% of the time based on the area under the ROC curve, and its predictions closely matched actual observed rates of deficiency. The researchers identified two ways to use the score depending on the clinical goal: a lower cut-off that catches almost all truly deficient individuals (high sensitivity, 87–93%), useful for ruling out deficiency, and a higher cut-off that more precisely identifies who is most likely deficient (balancing sensitivity around 62% with specificity around 80%), useful for deciding who actually needs a blood test. Importantly, the score performed similarly across different seasons and in people not taking vitamin D supplements, suggesting it may be broadly applicable within this clinical setting. This research suggests that a clinical scoring tool could help doctors decide who actually needs a blood vitamin D test rather than testing everyone, which could reduce unnecessary laboratory costs and focus resources on those most likely to be deficient. However, because the study population was predominantly female, attended specialized bone disease clinics in Italy, and was not validated in a separate external group, the score's usefulness in other populations — such as men, younger adults, or those seen in general practice — remains to be confirmed.

Check Your Own Numbers

Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.

Upload Your Labs

Have a question about this study?

Citation

Nuti R, Gennari L, Frediani B, Gonnelli S, Merlotti D, Caffarelli C, et al.. (2026). Development and Internal Validation of a Clinical Risk Score for Hypovitaminosis D in Italian Adults Aged &#x2265;50 Years Attending Osteoporosis and Metabolic Bone Disease Centers.. Nutrients. https://doi.org/10.3390/nu18172805