Association between magnesium depletion score and chronic obstructive pulmonary disease risk: a secondary data analysis from NHANES

Design

The NHANES database is a notable cross-sectional survey that provides nationally representative data on the health and nutrition of households in the USA.14 For our research study, we used data from the years 2001 to 2018. Prior to analysis, we excluded 57 546 cases with missing or incomplete COPD data, 6359 cases with missing or incomplete MDS data and 5050 cases with missing or incomplete information on other variables. Ultimately, our study included 39 852 participants (figure 1).

Figure 1

Flow chart of participants inclusion. NHANES, National Health and Nutrition Examination Survey.

Patient and public involvement

Patients and the public were not directly involved in the design, recruitment, or conduct of this study due to its retrospective nature using the NHANES database which includes de-identified data. However, we recognise the importance of patient and public involvement in research and plan to disseminate the results of this study through patient advocacy groups, public health outreach programmes and relevant online platforms to ensure that the findings are accessible to the broader community.

MDS

Based on the MDS model developed and validated by Fan et al, we quantified MDS using four relatively independent metrics.10 The first factor assessed was the use of diuretics. One point was attributed to individuals who had used a diuretic within the last 30 days, while those who had not received zero points. The second factor considered was the utilisation of proton pump inhibitors, which was consistent with diuretic use. One point was assigned to individuals using proton pump inhibitors. The third factor taken into account was the estimated glomerular filtration rate (eGFR). Participants were given one point if their eGFR was between 60 and 90, and two points if it fell below 60. Lastly, alcohol abuse was evaluated, and one point was given to male participants consuming more than two drinks per day, or female participants consuming more than one drink per day. We categorised the MDS into four classes: 0, 1, 2 and ≥3.

Dietary magnesium

We extracted data from the NHANES database, specifically using information from two dietary interviews conducted with participants.15 The first interview took place face-to-face at a Mobile Examination Center (MEC) and aimed to capture dietary intake over the previous 24 hours. Subsequently, a second interview was conducted via telephone within 3–10 days after the MEC visit. To minimise potential errors arising from unusual dietary activities, we calculated the average dietary magnesium (mg) intake for each participant.

COPD

We used the standardised diagnostic criteria recommended by the American Thoracic Society to evaluate the presence of COPD.16 Specifically, we assessed one-second force expiratory volume (FEV1) and forceful lung volume (FVC). Subsequently, we calculated the FEV1 to FVC ratio and considered an FEV1/FVC ratio of <0.7, in the presence of bronchodilator usage, as indicative of COPD.

Covariates

In our study, We included demographic data, lifestyle factors, medical history and haematologic marker information as covariates.

1. Demographic data: we considered age, sex, race and body mass index (BMI) as important demographic factors that could influence the association between MDS and COPD incidence.

2. Lifestyle factors: smoking, education and poverty income ratio (PIR) were included as lifestyle factors. These variables have been associated with both MDS and COPD risk, thus warranting consideration in our analysis.

3. Disease history: we assessed hypertension, diabetes and cardiovascular disease (CVD) as important disease history variables. Hypertension was defined as a mean systolic blood pressure of 140 mm Hg or higher, mean diastolic blood pressure of 90 mm Hg or higher, current use of hypertension medication or receiving a diagnosis of hypertension from a healthcare professional.17 Participants were classified as having diabetes if they had a fasting blood glucose level of 126 mg/dL or higher, glycosylated haemoglobin of 6.5% or higher, were taking insulin or glucose-lowering medications, or had been informed of their diabetes diagnosis.18 CVD was diagnosed if participants reported having coronary artery disease, heart failure, angina, heart attack or stroke.19

4. Haematologic markers: we included high-density lipoproteins and low-density lipoproteins as haematologic markers. These markers provide valuable information regarding participants’ lipid profile.

Statistical analysis

In our analysis, categorical variables were presented as frequencies and percentages, while continuous variables were reported as means and SD. Analysis of Variance (ANOVA) was used to analyse continuous variables and χ² test was used to analyse categorical variables.

The association between various variables and the risk of COPD was assessed using one-way logistic regression. We calculated the OR of MDS with COPD and their corresponding 95% CI through several adjusted models. To examine the robustness of the association between MDS and COPD risk across different populations, we conducted a stratified analysis. Additionally, we considered dietary magnesium intake and its potential impact on the association between MDS and COPD. Dietary magnesium intake was categorised into high and low groups based on the median value (264.5 mg).

All statistical analyses were performed using the weighting methodology outlined in the NHANES database analysis guidelines. We used R software (V.4.2.1) for the statistical analysis. A p value <0.05 was considered statistically significant.