This was a multicenter, prospective, non-interventional, observational study conducted at nine centers in India between 19 August 2020 and 25 June 2021 and involved adult patients diagnosed with vitamin D deficiency or insufficiency. The study investigators were either orthopedists, cardiologists, diabetologists, or general physicians. The study was conducted in compliance with the principles of the Declaration of Helsinki, International Conference on Harmonization - Good Clinical Practice (GCP) guidelines, and Indian regulatory guidelines (Indian Council of Medical Research and Indian GCP guidelines). The study protocol and protocol amendments, consent process, patient authorization form (PAF), the case report form (CRF), and all study-related documents were approved by the Royal Pune Independent Ethics Committee. All patients provided written informed consent prior to participating in any study-related procedures. The trial was prospectively registered on Clinical Trials Registry – India (CTRI/2020/03/024109) on 20 March 2020.

Each enrolled patient was prescribed vitamin D supplement as per physician’s discretion for 8 (non-diabetic patients) or 12 weeks (diabetic patients). The study consisted of a baseline visit on day 0, a follow-up visit at week 4, end-of study visit at week 8 for non-diabetic patients, and end-of-study visit at week 12 for diabetic patients. The total study duration was approximately 10 months including enrolment period. All patients were followed-up for primary and secondary endpoints.

Male or female patients aged ≥18 years diagnosed with vitamin D deficiency (serum 25 [OH] D level <20 ng/mL) or insufficiency (serum 25 [OH] D level of 21-29 ng/mL) of known or unknown origin who provided written informed consent were included in the study. In addition, patients with hypertension and diabetes were required to be on stable anti-hypertension and anti-diabetes treatment two months prior to enrollment to be eligible for inclusion. Patients who were on treatment for vitamin D deficiency or insufficiency, patients with serum 25(OH) D level >30 ng/mL, and pregnant and lactating women were excluded.

The primary endpoints were 1) to understand the patient profiles in terms of demographic and anthropometric measurements, exposure to sunlight, lifestyle pattern, and socioeconomic status, and 2) to determine the clinical characteristics in terms of clinical signs and symptoms at baseline and weeks 4 and 8. The key secondary endpoints were 1) to assess the type and number of comorbidities including their duration at baseline, 2) to assess the prescription pattern of vitamin D supplements by type of formulation, dose, route, and frequency of administration at weeks 4 and 8, 3) to assess change in 25(OH) D levels from baseline to weeks 4 and 8 and 4) to assess number and proportion of patients with adverse drug reactions (ADRs)/serious ADRs. The key exploratory endpoints were 1) to assess change in QoL score from baseline to week 8, 2) to assess mean change in systolic blood pressure (SBP) and diastolic blood pressure (DBP) from baseline to 8 weeks in hypertensive patients with no modification in antihypertensive medications during the study period and 3) to assess mean change in glycated hemoglobin (HbA1c) levels from baseline to 12 weeks in diabetic patients with no modification in antidiabetic medications during the study period.

Patient profiles

Patients were grouped by gender, age groups (18 to ≤30, 31 to ≤40, 41 to ≤50, or >50 years), exposure to sunlight, time of exposure to sun (morning, afternoon, or evening hours), past sunscreen use, daily frequency of past sunscreen use (once a day, twice a day, or more than thrice a day), current sunscreen use daily frequency of current sunscreen use (once a day, twice a day, or more than thrice a day), skin color (brown, olive, pale, dark brown, or reddish), dietary habits (vegetarian and non-vegetarian, non-vegetarian, or vegetarian), smoking habits (non-smoker, former, or current smoker), alcohol consumption habits (non-drinker, former or current drinker), education level (illiterate, primary school certificate, middle school certificate, high school certificate, intermediate/post-high school diploma, graduate/post-graduate, or professional/honors), occupation (unemployed, unskilled worker, semi-skilled worker, skilled worker, clerical, shop-owner, farmer, semi-professional, or professional), monthly family income in Indian rupees (≥2000, 1000-1999, 750-999, 500-749, 300-499, 101-299, or ≤100), socioeconomic status based on Kuppuswamy scale ¹⁵ (lower class [score <5], upper-lower class [score 5-10], lower-middle class [score 11-15], upper-middle class [score 16-25], or upper class [score 26-29]), for assessment of association with vitamin D deficiency or insufficiency.

Clinical characteristics

The 100-mm visual analog scale (VAS) was used to assess patients’ clinical signs and symptoms such as increased risk of illness/infections, excessive fatigue and tiredness, bone and lower back pain, depression, impaired/poor wound healing, bone loss or low bone mineral density, hair loss, muscle pain, back pain, dizziness, left knee pain, left sided subtruch, leg cramps, pain in heel, pain in lower limb, tingling in hand and feet, and weakness. VAS is a self-administered, continuous scale anchored by two verbal descriptors on either extreme, i.e., 0 mm representative of no symptoms and 100 mm representative of worst imaginable symptoms. Changes in VAS scores for each sign and symptom from baseline to weeks 4 and 8 were evaluated.

Comorbidities associated with vitamin D deficiency/insufficiency

Proportion of patients by number of comorbidities, their duration, and proportion by system organ class (SOC) and preferred term (PT) were assessed at baseline. SOCs and PTs were coded using the Medical Dictionary for Regulatory Activities MedDRA version 24.0.

Prescription patterns of vitamin D supplements

Patients received vitamin D supplements as per physician discretion. The formulation type, dose, route and frequency of use were assessed at weeks 4 and 8.

Changes in clinical biomarkers

Changes in vitamin D levels from baseline to weeks 4 and 8 were assessed for all patients. In addition, change in SBP and DBP of hypertensive patients from baseline to week 8, and change from in HbA1c levels of diabetic patients from baseline to week 12 were also evaluated.

Assessment of QoL

The 20-item short form survey (SF-20) was used to assess QoL at baseline and week 8. ¹⁶ This questionnaire consists of 20 questions grouped into six domains including physical functioning (PF: 6 questions), role problem (RP: 2 questions), social functioning (SF: 1 question), mental health (MH: 5 questions), current health perceptions (CH: 5 questions), and body pain (BP: 1 question). For all questions, except questions 1 and 3, the lowest raw score represented worst possible score and highest score represented best possible score. The raw scores for questions 1 and 3 were reversed to align with the scores of the remaining questions. The highest and lowest possible scores were 18 and 6 (range 12) for PF, 6 and 2 (range 4) for RP, 6 and 1 (range 5) for BP, 6 and 1 (range 5) for SF, 30 and 5 (range 25) for MH, and 25 and 5 (range 20) for CH, respectively. The scores were then transformed to the 0% to 100% scale such that transformed score = ([actual raw score – lowest possible raw]/raw score range) × 100. Scores across each of these domains were reported on a 0% to 100% scale, with 0% representing “the worst possible score” in that domain and 100% “the best possible score”. Patient’s QoL assessment was done at baseline and week 8 by using SF-20 questionnaire. ¹⁶ The transformed QoL scores and change in scores for each domain from baseline to week 8 were determined.

Safety and tolerability assessments

Incidence of adverse drug reactions (ADRs) and serious ADRs and their relationship to study treatment were monitored throughout the study. All ADRs and serious ADRs were classified along with the standards of Medical Dictionary for Regulatory Activities (MedDRA) version 24.0 (RRID:SCR_003751) and grouped by PT and SOC. ADR was defined as a response to a medicinal product that was noxious and unintended, where response meant that a causal relationship between the study drug and the adverse event was at least a reasonable possibility. A serious ADR was defined as an ADR that resulted in the death of a patient, was life threatening, required hospitalization, caused prolongation of hospitalization, caused persistent or significant disability/incapacity, or required medical or surgical intervention to prevent serious outcome.

As this was an observational study, no formal statistical sample size calculation was performed. A total of 201 patients were planned to be enrolled.

Enrolled population comprised of all patients who met the eligibility criteria and were enrolled in the study. The full analysis set (FAS) comprised of patients from the enrolled population with at least one post-baseline visit and had no major protocol violations that could have altered the integrity of the study.

25[OH]D) levels by patient profiles were summarized using descriptive statistics using the FAS population and statistical analysis was performed using analysis of variance (ANOVA) at 5% level of significance. Clinical characteristics as assessed by change in clinical signs and symptoms using VAS scores from baseline to weeks 4 and 8 were summarized as descriptive statistics using the categorical variables summarized as frequency counts (n) and percentages (%). Continuous variables were summarized as mean (standard deviation [SD]) and analyzed using paired t test unless otherwise specified. Data were analyzed using SAS ^® version 9.4 (SAS Institute Inc., USA; RRID:SCR_008567).

Of 248 screed patients, 201 met the eligibility criteria, were enrolled in the study and comprised the enrolled population ( Figure 1). Of these, 23 patients did not have at least one post-baseline visit and were excluded from the FAS. Of the 201 enrolled patients, 45 were lost to follow-up and 2 withdrew consent; thus, 154 patients completed the study. Table 1 summarizes the demographics and baseline characteristics of the enrolled population. The study had a female preponderance (61.7% women), with the mean (SD) age and BMI of the enrolled population being 43.4 (14.52) years and of 27.1 (5.52) kg/m ². Among the enrolled patients, 24.4% patients comprised the diabetes subgroup, 15.4% patients comprised the hypertension subgroup, and 8.9% patients had both diabetes and hypertension. There were 87 patients who had at least one comorbidity and the mean (SD) duration of comorbidities was 1.96 (3.127) years. Types of comorbidities observed in decreasing order of frequency were diabetes mellitus, hypertension, asthenia, arthralgia, back pain, fatigue, hypothyroidism, myalgia, anemia, thyroid disorder, gastritis, gait disturbance, pain, obesity, pain in extremity, peripheral neuropathy, and insomnia.

Baseline 25(OH) D levels were assessed by various demographic and anthropometric factors, and subgroups within these factors were assessed for significant association with poor vitamin D levels. Most of these factors did not influence vitamin D levels, except for current sunscreen use ( P = 0.0297) and socioeconomic status ( P = 0.0222; Table 2). Gender, age category, exposure to sunlight, time of exposure to sun, past sunscreen use, daily frequency of past sunscreen use, skin color, dietary habits, smoking habits, alcohol consumption habits, education level, occupation, family income, and socioeconomic status had no significant effect ( P > 0.05) on vitamin D status deficiency versus insufficiency. However, an odds ratio of 5.88 for patients in the upper-lower class indicated that these patients had 4.88-fold ( P = 0.033) significantly higher odds for vitamin D deficiency versus insufficiency in comparison with patients in the lower class ( Table 2).

Vitamin D supplementation and change in vitamin D levels

Enrolled patients were prescribed different vitamin D supplements such as 25-OH cholecalciferol, cholecalciferol, and ergocalciferol. Of 178 patients in the FAS, 140 (78.7%) were receiving cholecalciferol, 4 (2.2%) were receiving ergocalciferol at a dose of 60,000 IU/week orally, and 2 (1.1%) were receiving 25-OH cholecalciferol at a dose of 60000 IU/week orally, at week 4. Among the 140 (78.7%) patients on cholecalciferol, 134 (75.3%) were receiving 60000 IU and 6 (3.4%) were receiving 400 IU. For these patients, the route was oral for 132 (74.2%) patients and intramuscular for 8 (4.5%) patients. Frequency was daily for 6 (3.4%) patients and once a week for 134 (75.3%) patients.

At week 8, 147 (82.6%) patients were receiving cholecalciferol, of which 141 (79.2%) patients were receiving 60,000 IU and 6 (3.4%) patients were receiving 400 IU; of these, the route was oral for 139 (78.1%) patients and intramuscular for 8 (4.5%) patients. Frequency was daily for 6 (3.4%) patients and once a week for 141 (79.2%) patients. Ergocalciferol was prescribed to 11 (6.2%) patients at a dose of 60,000 IU/week.

The mean (SD) serum 25(OH) D levels were increased from baseline value of 14.57 (6.839) ng/mL to 27.86 (16.629) ng/mL at week 4 and 31.94 (28.995) ng/mL at week 8 ( Table 3). The change from baseline values was statistically significant at all both-baseline visits (week 4: 13.25 (16.605), P < 0.0001; week 8: 17.57 (28.186), P < 0.0001).

Change in clinical signs and symptoms

Changes in VAS scores of clinical signs/symptoms from baseline to weeks 4 and 8 are shown in Figure 2. Very few patients experienced back pain, dizziness, left knee pain, left sided subtruch, leg cramps, pain in heel, pain in lower limb, tingling in hand and feet, or weakness at baseline; hence, these symptoms were not analyzed for changes at weeks 4 and 8. Among the remaining symptoms, the mean (SD) VAS scores of bone and lower back pain, bone loss/low bone mineral density, and muscle pain were significantly reduced from baseline to week 4 ( P < 0.0001 for all) and week 8 ( P < 0.0001 for all, except muscle pain [ P = 0.0009]) VAS scores for risk of illness/infections and excessive fatigue and tiredness also tended to decrease, but these changes were not significantly significant ( Figure 2).

Effect of vitamin D supplementation on blood pressure of hypertensive patients

Among patients with hypertension, i.e. patients who were on stable antihypertension medication for two months before study enrollment, there was statistically significant decrease in the mean (SD) SBP from 120.0 (15.63) mmHg at baseline to 111.2 (8.16) mmHg at week 8 ( P = 0.0212) and in mean DBP from 84.8 mmHg at baseline to 79.2 mmHg at week 8 ( P = 0.0110) ( Table 3).

Effect of vitamin D supplementation on HbA1c levels of diabetic patients

Among patients with diabetes, i.e., patients who were on stable anti-diabetes medications for two months prior to enrollment, the mean HbA1c decreased significantly from baseline (8.11%) to week 12 (6.49%) ( P = 0.0143) ( Table 3).

Effect of vitamin D supplementation on quality of life scores

Out of the six domains comprising the SF-20 questionnaire for (QoL) assessment, improvement was observed in four domains after two months of vitamin D supplementation. There was an improvement in mean (SD) score of 10.92 (46.259) points for physical functioning, 21.52 (43.482) points for role functioning, 35.42 (38.299) for body pain, and 9.15 (18.137) points for health perception from baseline to week 8 ( Figure 3).

No ADRs, serious ADRs, or OPRIs were reported after two months of vitamin D supplementation. All the patients had normal results for vital parameters (pulse rate, respiratory rate, body temperature and blood pressure) and physical examination after two months of vitamin D supplementation (data not shown). None of the patients discontinued the vitamin D treatment due to any ADR. At the end of two months, 2 patients had 25[OH] D levels >100 ng/mL and 2 patients had levels >150 ng/mL.