Skip to main content
NCBI home page
Preventive Medicine Reports logoLink to Preventive Medicine Reports
. 2025 May 5;54:103101. doi: 10.1016/j.pmedr.2025.103101

Association of hemoglobin, albumin, lymphocyte, and platelet score with all-cause and cardiovascular mortality in adults with diabetes or prediabetes: Results from NHANES 2005–2018

Taotao Zhang 1,⁎, Peiqian Liu 1
PMCID: PMC12136913  PMID: 40469253

Abstract

Objective

The relationship between the hemoglobin, albumin, lymphocyte, and platelet (HALP) score and mortality in patients with diabetes/prediabetes remains uncertain. This study aimed to evaluate linear and non-linear associations between the HALP score and all-cause and cardiovascular mortality in this population and identify potential clinically relevant thresholds.

Methods

We analyzed data from 19,350 United States adults with diabetes/prediabetes using two linked datasets: the National Health and Nutrition Examination Survey (NHANES, 2005–2018) and mortality records from the National Death Index (NDI), with follow-up through December 31, 2019. Kaplan-Meier survival curves, Cox proportional hazards models, and restricted cubic splines (RCS) were used to evaluate the HALP score and mortality associations.

Results

Kaplan-Meier analysis revealed the highest all-cause mortality in Q1, the lowest all-cause mortality in Q3, and the lowest cardiovascular mortality in Q4 (p-value <0.0001). Cox regression analysis demonstrated significantly reduced risks of all-cause mortality (HR 0.64, 95 % CI 0.58, 0.73) and cardiovascular mortality (HR 0.58, 95 % CI 0.42, 0.82) in Q4 compared to Q1. RCS identified an L-shaped association between the HALP score and mortality, with inflection points at 42.29 (all-cause) and 39.98 (cardiovascular).

Conclusion

An L-shaped association between the HALP score and both all-cause and cardiovascular mortality in participants with diabetes mellitus or prediabetes.

Keywords: Hemoglobin, albumin, lymphocyte, and platelet score; All-cause mortality; Cardiovascular mortality; Diabetes; Prediabetes

Highlights

  • •

    Nutrition-inflammation score links to mortality in dysglycemia.

  • •

    Composite score shows L-curve mortality association.

  • •

    42 (all-cause) and 40 (cardiac) points define death risk thresholds.

  • •

    Top scorers cut all-cause death 36 %, cardiac mortality 42 %.

1. Introduction

In recent decades, diabetes and its complications have incurred significant global health and economic detriment(Matoori, 2022). According to the International Diabetes Federation, the global prevalence of diabetes among adults aged 20–79 reached 10.5 % (536.6 million cases) in 2021, with projections indicating a rise to 12.2 % (783.2 million cases) by 2045(Sun et al., 2022). The increase in the prevalence of diabetes is closely linked to the increase in the number of people with prediabetes. Prediabetes represents an intermediate metabolic state between normal glucose homeostasis and diabetes, characterized by elevated blood glucose levels that fall below the diagnostic threshold for diabetes. Previous research has demonstrated that individuals with diabetes/prediabetes have an increased risk of cardiovascular disease (CVD), renal dysfunction, peripheral vascular disease, and diabetic retinopathy(Lan et al., 2024; Marassi and Fadini, 2023; Zhou et al., 2024). In addition, cardiovascular mortality and all-cause mortality are significantly elevated in individuals with diabetes compared to non-diabetic populations(Raghavan et al., 2019). Therefore, identifying additional risk factors is essential for enhancing prognostic outcomes in high-risk populations.

The hemoglobin, albumin, lymphocyte, and platelet (HALP) score was proposed as an emerging biomarker and was initially applied by CHEN X L et al. to predict survival outcomes in gastric cancer patients(Chen et al., 2015). Since then, the HALP score has been used mainly in prognostic studies of various types of cancer(Zhang et al., 2023a; Zhang et al., 2023b; Zhou and Yang, 2023). The HALP score combines immune, inflammatory, and nutritional indicators. It effectively evaluates the body's inflammation levels and nutritional balance through a single measurement. Research has demonstrated that inflammatory indices derived from lymphocyte-related parameters (e.g., neutrophil-lymphocyte ratio, NLR; monocyte-lymphocyte ratio, MLR; systemic immune-inflammatory index) are linked to diabetes-related mortality(Dong et al., 2023; Li et al., 2024; Yang et al., 2024). LEE G et al. found that lower or higher hemoglobin concentrations were shown to be risk factors for cardiovascular and all-cause mortality in diabetic and non-diabetic patients(Lee et al., 2018). Lower serum albumin levels have been correlated with coronary calcification, adverse vascular events, and all-cause mortality(Seidu et al., 2020; Won et al., 2024). Multiple large-scale cohort studies have established associations between the HALP score and mortality in patients with coronary artery disease, the general population, and individuals with osteoarthritis(Li et al., 2025; Pan and Lin, 2023; Zheng et al., 2023). However, the association between the HALP score and mortality risk remains uninvestigated in individuals with diabetes/prediabetes. Given that individual HALP components (e.g., hemoglobin and albumin) exhibit non-linear relationships with mortality in chronic diseases(Lee et al., 2018; Seidu et al., 2020; Won et al., 2024), we hypothesized that the HALP score itself might demonstrate non-linear associations with mortality, reflecting potential saturation or threshold effects inherent to its constituent biomarkers.

Therefore, this study analyzed National Health and Nutrition Examination Survey (NHANES) data to evaluate both linear and non-linear associations between the HALP score and all-cause and cardiovascular mortality in individuals with diabetes/prediabetes, and to identify potential clinically relevant thresholds.

2. Methods

2.1. Data sources

The NHANES is a nationally representative survey conducted by the National Center for Health Statistics (NCHS), employing a complex, stratified, multistage probability sampling design. Data collection involved three primary methods: household interviews, standardized physical examinations at mobile examination centers, and laboratory tests. All NHANES protocols received ethical approval from the NCHS, and written informed consent was obtained from all participants.

Our analysis integrated two complementary data components. First, NHANES baseline data included seven consecutive biennial cycles (2005–2018), covering 67,203 participants. The dataset encompassed demographic characteristics, clinical measurements, and laboratory test results, which were used to define exposure variables and covariates. Second, mortality data were linked by the NCHS through probabilistic matching of NHANES participants with the National Death Index (NDI), with follow-up through December 31, 2019. The matching algorithm utilized encrypted identifiers, including Social Security Number, name, and date of birth, and a unique participant sequence number (SEQN) ensured precise one-to-one linkage between baseline data and mortality records.

2.2. Study population

From the initial NHANES cohort (2005–2018; n = 67,203), we applied sequential exclusion criteria. First, we excluded participants aged <20 years (n = 27,454). Second, we excluded those with missing data: 4,156 participants lacked laboratory values (platelet, lymphocyte, or albumin), and 71 lacked mortality linkage data. Third, participants were classified into diabetes/prediabetes groups using criteria from references (Diagnosis and Classification of Diabetes, 2025; Bergman et al., 2024). Diabetes was defined as meeting one or more of the following: fasting glucose ≥7.0 mmol/L, random glucose or oral glucose tolerance test (OGTT) 2-h blood glucose ≥11.1 mmol/L, hemoglobin A1c (HbA1c) ≥6.5 %, use of insulin or oral hypoglycemic agents, or self-reported physician-diagnosed diabetes (“Has a doctor ever told you that you have diabetes?”). Prediabetes was defined as meeting one or more of: fasting glucose 6.1–6.9 mmol/L, OGTT 2-h blood glucose 7.8–11.0 mmol/L, HbA1c 5.7–6.4 %, or self-reported physician-diagnosed prediabetes (“Has a doctor ever told you that you have prediabetes or borderline diabetes?”). Participants not meeting criteria for diabetes/prediabetes (n = 16,172) were excluded. The final analytic cohort included 19,350 individuals (6667 with diabetes; 12,683 with prediabetes).

2.3. Calculating the HALP score

Quantitative laboratory parameters (lymphocyte count, hemoglobin concentration, platelet count, serum albumin level) were derived from complete blood count and standardized biochemical assays in NHANES. The HALP score was calculated as: HALP score = [lymphocytes (/L) × hemoglobin (g/L) × albumin (g/L)]/platelets (/L)(Ding et al., 2024). Participants were stratified into quartiles (Q1–Q4) based on the HALP score values, where Q1 corresponded to the lowest 25 % and Q4 to the highest 25 %. The HALP score was standardized via Z-score transformation to adjust for interindividual variability in baseline biomarker levels.

2.4. Mortality assessment

First, follow-up time was calculated in person-months from the date of each participant's NHANES interview to the earliest of the following events: (1) date of death (accurate to the month) or (2) the study endpoint (December 31, 2019). Participants who remained alive at the endpoint were right-censored in survival analyses. Second, mortality outcomes were defined as follows: (1) all-cause mortality encompassed all death events recorded in the NDI, regardless of cause; (2) cardiovascular mortality was identified based on the underlying cause of death in NDI records, classified using the International Classification of Diseases, Tenth Revision (ICD-10) codes I00-I09, I11, I13, I20-I51, and I60-I69, consistent with the Global Burden of Disease study standards for CVD classification.

2.5. Covariates

By extant literature, the following covariates were included: age, sex, race/ethnicity, education level, marital status, household income, smoking status, drinking status, CVD, hypertension, body mass index (BMI), total cholesterol, and high-density lipoprotein (HDL)(Chen et al., 2024a; Liu and Liang, 2024; Yu et al., 2024; Zhao et al., 2024). Medication use (e.g., statins, metformin) was excluded from covariate adjustment due to incomplete dosage and adherence data in NHANES. Improper modeling of these incomplete variables could lead to residual confounding. The race/ethnicity categories encompass Non-Hispanic White, Non-Hispanic Black, Mexican American, Other Hispanic, and Other Race. Educational level is categorized into three levels based on the number of years of schooling: <9 years, 9–12 years, and > 12 years. Marital status is categorized as follows: living alone (unmarried, separated, divorced, widowed) and living with a partner (married, living with a partner). Household income was stratified into three tiers using the poverty-to-income ratio (PIR): low (PIR ≤1.3), middle (PIR 1.3–3.5), and high (PIR >3.5). Smoking status was stratified as: never smokers (<100 lifetime cigarettes), former smokers (≥100 cigarettes with cessation), and current smokers (≥100 cigarettes with continued use). Alcohol consumption was defined as never drinkers (<12 lifetime drinks), former drinkers (≥12 drinks in any prior year with ≥1-year abstinence), and current drinkers (≥12 drinks annually with consumption within the past year). CVD history was defined by self-reported diagnoses of coronary heart disease, angina pectoris, stroke, myocardial infarction, or congestive heart failure. Confirmation required meeting ≥1 diagnostic criterion within this classification system. Hypertension was defined by self-reported physician diagnosis. BMI was calculated via anthropometric measurements and categorized per WHO criteria: underweight or normal (<25.0 kg/m2), overweight (25.0–29.9 kg/m2), and obese (≥30.0 kg/m2). Total cholesterol and HDL were extracted from standardized biochemistry laboratory data.

2.6. Statistical analyses

Continuous variables were analyzed using one-way ANOVA and expressed as mean ± standard deviation (SD). Categorical variables were analyzed using chi-square tests and expressed as frequencies with percentages (n (%)). The incidence of all-cause mortality and cardiovascular mortality was calculated during the follow-up period. Kaplan-Meier survival curves were used to explore differences in participant survival between the four groups. Multivariable Cox proportional hazards models assessed associations between the HALP score and mortality outcomes. Hazard ratios (HR) with 95 % confidence intervals (CI) were calculated for all-cause and cardiovascular mortality. The HALP score was analyzed both as a categorical variable (divided into quartiles: Q1-Q4) and as a standardized continuous variable (per 1-SD increment). Proportional hazards (PH) assumptions were tested using Schoenfeld residuals. Global and covariate-specific tests were performed for all variables in the final Cox model. A non-significant P-value (p-value >0.05) indicated no violation of the PH assumption. Two analytical models were constructed: Model 1 (unadjusted); Model 2 (fully adjusted: age, sex, race/ethnicity, education level, marital status, household income, smoking status, drinking status, CVD, hypertension, BMI, total cholesterol, and HDL). Missing data were addressed via multivariable imputation. Given the potential for nonlinear relationships between composite biomarkers and mortality, restricted cubic splines (RCS) analyses were preplanned to evaluate dose-response patterns and identify clinically relevant thresholds. RCS regression with four knots at the 5th, 35th, 65th, and 95th percentiles of the HALP score was applied after adjusting for covariates (as in Model 2), evaluating the dose-response relationship between the HALP score and mortality. A nonlinear association was confirmed if the p-value for nonlinearity was <0.05. To identify potential threshold effects, a smoothed two-piece Cox regression model was used to analyze the association between the HALP score and mortality. The optimal inflection point was identified by comparing nested linear and threshold models via likelihood ratio tests (p-value <0.05 indicating improved fit), with the final threshold determined through bootstrap resampling. Finally, we performed sensitivity analyses to assess the robustness of the results. First, to minimize the possibility of reverse causality, we excluded participants with less than 24 months of follow-up to delay entry into the study population. Second, to minimize the effect of extreme values, we excluded data with the HALP score less than Q1–1.5 * interquartile range(IQR), or greater than Q3 + 1.5 * IQR. Third, as the HALP score is often associated with prognosis in patients with cancer and cancer is a leading cause of death in patients with diabetes, we excluded participants who self-reported a cancer diagnosis at baseline to reduce potential confounding. Fourth, renal dysfunction may affect albumin levels. We excluded participants with chronic kidney disease (CKD) (estimated glomerular filtration rate < 60 mL/min/1.73m2) at baseline. All statistical analyses were performed using Free Statistics Software (version 2.0; Beijing, China, http://www.clinicalscientists.cn/freestatistics). A two-tailed p-value of <0.05 is considered statistically significant.

3. Results

3.1. Baseline characteristics of participants

This study analyzed 19,350 U.S. adults with diabetes/prediabetes stratified by HALP score quartiles. The lowest HALP group (Q1: 4.59–37.81) was older (60.6 ± 15.5 vs. Q4: 56.4 ± 15.3; p-value<0.001), had more women (62.8 % vs. 34.1 %), and a higher proportion of non-Hispanic Whites (40.9 % vs. 36.7 %), while Mexican Americans were more concentrated in Q4 (19 % vs. 14.5 %; p-value<0.001). Education showed statistical differences (p-value = 0.005) but minimal actual variation without a clear gradient. Q1 had higher middle-income households (40.3 % vs. 38.4 %; p-value<0.001) and more people living alone (45.6 % vs. 36.0 %; p-value<0.001). Q1 had the highest never-smokers (54.7 % vs. 45.0 %) but lowest current drinkers (55.7 % vs. 64.9 %; p-value<0.001). Despite similar obesity rates (49–50 %), Q1 had higher hypertension (65.3 % vs. 57.7 %) and CVD prevalence (22.4 % vs. 17.9 %; p-value<0.001). Total cholesterol increased with higher HALP (Q1: 4.9 ± 1.1 mmol/L vs. Q4: 5.1 ± 1.2; p-value<0.001) while HDL-C decreased (Q1: 1.4 ± 0.4 vs. Q4: 1.2 ± 0.4; p-value<0.001). Prediabetes prevalence peaked in Q3 (67.4 % vs. Q1: 63.8 %; p-value = 0.002) (See Table 1).

Table 1.

Descriptive characteristics of United States adults aged 20 years or older with diabetes or prediabetes: Data from the NHANES, 2005–2018.

Variables Total Q1
(4.59–37.81)		 Q2
(37.82–50.31)		 Q3
(50.32–66.22)		 Q4
(66.23–5516.55)		 *P value
Number 19,350 4838 4837 4837 4838
Age, Mean ± SD 58.2 ± 15.4 60.6 ± 15.5 58.6 ± 15.2 57.0 ± 15.2 56.4 ± 15.3 < 0.001
Sex, n (%) < 0.001
 Male 9752 (50.4) 1801 (37.2) 2172 (44.9) 2593 (53.6) 3186 (65.9)
 Female 9598 (49.6) 3037 (62.8) 2665 (55.1) 2244 (46.4) 1652 (34.1)
Race/ethnicity, n (%) < 0.001
 Non-Hispanic White 7405 (38.3) 1981 (40.9) 1858 (38.4) 1789 (37) 1777 (36.7)
 Non-Hispanic Black 4645 (24.0) 1372 (28.4) 1143 (23.6) 1082 (22.4) 1048 (21.7)
 Mexican American 3231 (16.7) 700 (14.5) 750 (15.5) 861 (17.8) 920 (19)
 Other Hispanic 1947 (10.1) 371 (7.7) 533 (11) 532 (11) 511 (10.6)
 Other Race 2122 (11.0) 414 (8.6) 553 (11.4) 573 (11.8) 582 (12)
Education level, n (%) 0.005
 < 9 5880 (30.4) 1487 (30.7) 1395 (28.8) 1475 (30.5) 1523 (31.5)
 9–12 4565 (23.6) 1111 (23) 1114 (23) 1154 (23.9) 1186 (24.5)
 >12 8905 (46.0) 2240 (46.3) 2328 (48.1) 2208 (45.6) 2129 (44)
Marital status, n (%) < 0.001
 Living with a partner 11,691 (60.4) 2633 (54.4) 2926 (60.5) 3037 (62.8) 3095 (64)
 Living alone 7659 (39.6) 2205 (45.6) 1911 (39.5) 1800 (37.2) 1743 (36)
Household income, n (%) < 0.001
 Low 6518 (33.7) 1661 (34.3) 1585 (32.8) 1539 (31.8) 1733 (35.8)
 Middle 7620 (39.4) 1952 (40.3) 1908 (39.4) 1900 (39.3) 1860 (38.4)
 High 5212 (26.9) 1225 (25.3) 1344 (27.8) 1398 (28.9) 1245 (25.7)
Smoking status, n (%) < 0.001
 Never 9986 (51.6) 2648 (54.7) 2728 (56.4) 2435 (50.3) 2175 (45)
 Current 5781 (29.9) 1577 (32.6) 1389 (28.7) 1454 (30.1) 1361 (28.1)
 Former 3583 (18.5) 613 (12.7) 720 (14.9) 948 (19.6) 1302 (26.9)
Drinking status, n (%) < 0.001
 Never 3301 (17.1) 980 (20.3) 900 (18.6) 735 (15.2) 686 (14.2)
 Former 4156 (21.5) 1163 (24) 955 (19.7) 1027 (21.2) 1011 (20.9)
 Current 11,893(61.4) 2695(55.7) 2982(61.7) 3075(63.6) 3141(64.9)
Cardiovascular disease, n (%) < 0.001
 No 15,746 (81.4) 3754 (77.6) 3985 (82.4) 4033 (83.4) 3974 (82.1)
 Yes 3604 (18.6) 1084 (22.4) 852 (17.6) 804 (16.6) 864 (17.9)
Hypertension, n (%) < 0.001
 No 7592 (39.2) 1679 (34.7) 1886 (39) 1982 (41) 2045 (42.3)
 Yes 11,758 (60.8) 3159 (65.3) 2951 (61) 2855 (59) 2793 (57.7)
Body mass index, n (%) < 0.001
 Underweight or normal 3552 (18.4) 1034 (21.4) 934 (19.3) 834 (17.2) 750 (15.5)
 Overweight 6226 (32.2) 1394 (28.8) 1536 (31.8) 1612 (33.3) 1684 (34.8)
 Obese 9572 (49.5) 2410 (49.8) 2367 (48.9) 2391 (49.4) 2404 (49.7)
Total cholesterol, Mean ± SD 5.0 ± 1.1 4.9 ± 1.1 5.0 ± 1.1 5.0 ± 1.1 5.1 ± 1.2 < 0.001
High-density lipoprotein, Mean ± SD 1.3 ± 0.4 1.4 ± 0.4 1.3 ± 0.4 1.3 ± 0.4 1.2 ± 0.4 < 0.001
Diabetes, n (%) 6667 (34.5) 1750 (36.2) 1636 (33.8) 1579 (32.6) 1702 (35.2) 0.002
Prediabetes, n (%) 12,683 (65.5) 3088 (63.8) 3201 (66.2) 3258 (67.4) 3136 (64.8) 0.002
Open in a new tab

Note: Continuous variables are presented as mean ± standard deviation (SD) and analyzed using one-way ANOVA. Categorical variables are expressed as frequencies with percentages (n (%)) and analyzed using chi-square tests. p values <0.05 were considered statistically significant.

Abbreviations: NHANES, National Health and Nutrition Examination Survey.

3.2. Association of the HALP score with all-cause mortality and cardiovascular mortality

Over a mean 84.2-month follow-up, 3129 all-cause deaths (16.17 %) and 834 cardiovascular deaths (4.31 %) were recorded in the cohort. Kaplan-Meier analysis revealed significant mortality differences across quartiles (p-value<0.0001). Q1 exhibited the highest mortality, while Q3 demonstrated the lowest all-cause mortality. Cardiovascular mortality was minimized in Q4 (Fig. 1A and B).

Fig. 1.

Fig. 1

Open in a new tab

Kaplan-Meier analysis of the hemoglobin, albumin, lymphocyte, and platelet score in relation to all-cause (1A) and cardiovascular mortality (1B) among United States adults aged 20 years or older with diabetes or prediabetes: Data from the NHANES, 2005–2018.

Abbreviations: NHANES, National Health and Nutrition Examination Survey; Q1–Q4 refer to the first to fourth quartiles of the hemoglobin, albumin, lymphocyte, and platelet score, with ranges of Q1 (4.59–37.81), Q2 (37.82–50.31), Q3 (50.32–66.22), and Q4 (66.23–5516.55).

Table 2 summarizes Cox regression analyses of HALP associations with mortality. For all-cause mortality, compared to Q1, Q4 showed 45 % lower risk in Model 1 (unadjusted; HR 0.55, 95 % CI 0.50, 0.61), persisting as 36 % lower risk after full adjustment (Model 2; HR 0.64, 95 % CI 0.58, 0.71). Each HALP SD increase was associated with 22 % (HR 0.78, 95 % CI 0.75, 0.82) and 15 % (HR 0.85, 95 % CI 0.82, 0.89) lower risks in Models 1 and 2, respectively. For cardiovascular mortality, compared to Q1, Q4 demonstrated 59 % lower risk in Model 1 (HR 0.41, 95 % CI 0.34, 0.50), remaining 42 % lower risk in Model 2 (HR 0.58, 95 % CI 0.42, 0.82). Each HALP SD increase corresponded to 63 % (HR 0.37, 95 % CI 0.29, 0.48) and 49 % (HR 0.51, 95 % CI 0.40, 0.64) reduced risks in Models 1 and 2. (See Table 2.)

Table 2.

Association of the hemoglobin, albumin, lymphocyte, and platelet score with all-cause and cardiovascular mortality among United States adults aged 20 years or older with diabetes or prediabetes: Cox regression analysis from the NHANES, 2005–2018.

Exposure
Model 1
Model 2
HR (95 % CI) HR (95 % CI)
All-cause death
Hemoglobin, albumin, lymphocyte, and platelet score
 Q1 (4.59–37.81) 1.00 1.00
 Q2 (37.82–50.31) 0.59 (0.53, 0.64) 0.64 (0.58, 0.70)
 Q3 (50.32–66.22) 0.53 (0.48, 0.59) 0.65 (0.59, 0.72)
 Q4 (66.23–5516.55) 0.55 (0.50, 0.61) 0.64 (0.58, 0.71)
Per standard deviation increase 0.78 (0.75, 0.82) 0.85 (0.82, 0.89)
Cardiovascular death
Hemoglobin, albumin, lymphocyte, and platelet score
 Q1 (4.59–37.81) 1.00 1.00
 Q2 (37.82–50.31) 0.42 (0.35, 0.51) 0.47 (0.39, 0.58)
 Q3 (50.32–66.22) 0.47 (0.40, 0.57) 0.64 (0.51, 0.81)
 Q4 (66.23–5516.55) 0.41 (0.34, 0.50) 0.58 (0.42, 0.82)
Per standard deviation increase 0.37 (0.29, 0.48) 0.51 (0.40, 0.64)
Open in a new tab

Note: Model 1 (unadjusted); Model 2 (fully adjusted: age, sex, race/ethnicity, education level, marital status, household income, smoking status, drinking status, cardiovascular disease, hypertension, body mass index, total cholesterol, and high-density lipoprotein).

Abbreviations: HR, hazard ratios; CI, confidence interval; NHANES, National Health and Nutrition Examination Survey; Q1–Q4, first to fourth quartiles of the hemoglobin, albumin, lymphocyte, and platelet score; Per standard deviation increase, per 1-standard deviation increase in the hemoglobin, albumin, lymphocyte, and platelet score.

3.3. Non-linear relationship between the HALP score and mortality

RCS Cox regression assessed nonlinear dose-response relationships between the HALP score and mortality in individuals with diabetes/prediabetes. The RCS results indicated that, after adjusting for all variables (as in Model 2), there was an L-shaped relationship between the HALP score and all-cause mortality (p for nonlinearity <0.05; Fig. 2A); HR decreased rapidly when the HALP score was <42.29 and slowly when the HALP score ≥ 42.29. Similarly, there was an L-shaped relationship between the HALP score and cardiovascular mortality (p for nonlinearity <0.05; Fig. 2B); HR decreased rapidly when the HALP score was <39.98 and changed slowly when the HALP ≥39.98.

Fig. 2.

Fig. 2

Open in a new tab

Restricted cubic spline analysis of the association between thehemoglobin, albumin, lymphocyte, and platelet score and all-cause mortality (2A) and cardiovascular mortality (2B) among United States adults aged 20 years or older with diabetes or prediabetes: Data from the NHANES, 2005–2018.

Note: Adjusted for age, sex, race/ethnicity, education level, marital status, household income, smoking status, drinking status, cardiovascular disease, hypertension, body mass index, total cholesterol, and high-density lipoprotein. The solid red line and pink area represent the hazard ratios and its corresponding 95 % confidence interval, respectively.

Abbreviations: NHANES, National Health and Nutrition Examination Survey. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Threshold analyses (Model 2-adjusted) revealed two inflection points. For all-cause mortality with the HALP score < 42.29, each 1-unit increase conferred a 3 % risk reduction (HR 0.97, 95 % CI 0.96, 0.97), with no association above this threshold (p-value ≥0.05). Similarly, each 1-unit increase in the HALP score < 39.98 conferred a 4 % reduction in cardiovascular mortality risk (HR 0.96, 95 % CI 0.95, 0.97), with no significant association observed at scores ≥39.98 (p-value ≥0.05) (See Table 3).

Table 3.

Threshold effect analysis of the hemoglobin, albumin, lymphocyte, and platelet score on all-cause and cardiovascular mortality among United States adults aged 20 years or older with diabetes or prediabetes: Data from the NHANES, 2005–2018.

Exposure Adjusted HR (95 %CI)
All-cause mortality
Hemoglobin, albumin, lymphocyte, and platelet score
 Inflection point 42.29
 <42.29 0.97 (0.96, 0.97)
 ≥42.29 0.99 (0.99, 1.00)
 Likelihood Ratio test p value <0.001
Cardiovascular mortality
Hemoglobin, albumin, lymphocyte, and platelet score
 Inflection point 39.98
 <39.98 0.96 (0.95, 0.97)
 ≥39.98 0.99 (0.99, 1.00)
 Likelihood Ratio test p value <0.001
Open in a new tab

Note: Both the all-cause mortality model and the cardiovascular mortality model adjusted for: age, sex, race/ethnicity, education level, marital status, household income, smoking status, drinking status, cardiovascular disease, hypertension, body mass index, total cholesterol, and high-density lipoprotein. Abbreviations: HR, hazard ratios; CI, confidence interval; NHANES, National Health and Nutrition Examination Survey.

The nonlinear associations between the HALP score and both all-cause and cardiovascular mortality were assessed in individuals with diabetes/prediabetes. An L-shaped association between the HALP score and both all-cause and cardiovascular mortality was observed in the diabetic population (p for nonlinearity <0.05 for both; Fig. 3A and B). Similarly, an L-shaped association between the HALP score and both all-cause and cardiovascular mortality was observed in the prediabetic population (p for nonlinearity <0.05 for both; Fig. 3C and D).

Fig. 3.

Fig. 3

Open in a new tab

Restricted cubic spline analysis of the association between the hemoglobin, albumin, lymphocyte, and platelet score and all-cause (3A) and cardiovascular mortality (3B) among United States adults aged 20 years or older with diabetes, and all-cause (3C) and cardiovascular mortality (3D) among United States adults aged 20 years or older with prediabetes: Data from the NHANES, 2005–2018.

Note: Adjusted for age, sex, race/ethnicity, education level, marital status, household income, smoking status, drinking status, cardiovascular disease, hypertension, body mass index, total cholesterol, and high-density lipoprotein. The solid red line and pink area represent the hazard ratios and its corresponding 95 % confidence interval, respectively.

Abbreviations: NHANES, National Health and Nutrition Examination Survey. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3.4. Sensitivity analyses

All sensitivity analyses used fully adjusted Cox models (Model 2). First, excluding participants with <24-month follow-up (n = 17,209), Q4 versus Q1 showed 26 % lower all-cause mortality (HR 0.74; 95 % CI 0.66, 0.82) and 48 % lower cardiovascular mortality (HR 0.52; 95 % CI 0.42, 0.65). Second, after excluding outliers (n = 18,795), Q4 had 37 % lower all-cause mortality (HR 0.63; 95 % CI 0.57, 0.70) and 50 % lower cardiovascular mortality (HR 0.50; 95 % CI 0.40, 0.61). Third, excluding cancer patients (n = 16,934), Q4 demonstrated 40 % lower all-cause mortality (HR 0.60; 95 % CI 0.53, 0.69) and 52 % lower cardiovascular mortality (HR 0.48; 95 % CI 0.38, 0.61). Fourth, excluding CKD patients (n = 16,190), Q4 exhibited 32 % lower all-cause mortality (HR 0.68; 95 % CI 0.59, 0.79) and 50 % lower cardiovascular mortality (HR 0.50; 95 % CI 0.37, 0.67; Supplementary Table 1).

4. Discussion

This study is the first to investigate the association between the HALP score and all-cause and cardiovascular mortality in patients with diabetes/prediabetes using a large NHANES-based sample. The main findings are as follows. First, the HALP score was independently associated with all-cause and cardiovascular mortality after adjusting for confounders. Second, the HALP score exhibited an L-shaped relationship with both all-cause mortality and cardiovascular mortality, with respective thresholds of 42.29 and 39.98. Third, sensitivity analyses demonstrated the robustness of our findings.

Prior studies have demonstrated that the HALP score is independently linked to retinopathy, erectile dysfunction, and post-stroke cognitive impairment(Chen et al., 2024a; Ding et al., 2024; Xu et al., 2023b). The HALP score has been used to predict clinical outcomes and is associated with a broad spectrum of conditions. In patients with coronary heart disease, the HALP score was negatively associated with the risk of all-cause mortality, a finding that was stable regardless of the presence or absence of heart failure(Zheng et al., 2023).In patients with heart failure, a high HALP score was associated with a reduced risk of death at 4, 12, 24 weeks, and 1 year following heart failure(Liu et al., 2024). Meta-analyses show that a low HALP score is associated with reduced overall and cancer-specific survival in cancer patients(Xu et al., 2023a). The above results suggest that the HALP score may serve as a clinical indicator for prognosis. The HALP score complements established inflammatory markers like NLR and PLR by integrating nutritional (albumin), immune (lymphocyte), and thrombotic (platelet) pathways. While NLR >3.48 predicts elevated mortality in diabetes (HR 2.03 for all-cause death) (Dong et al., 2023) and high PLR associates with 1.33-fold mortality risk in general populations(Mathur et al., 2019), the HALP score uniquely identifies an L-shaped risk threshold (e.g., 40 % lower mortality in Q4 vs. Q1, HR 0.60). The nonlinear HALP score-mortality relationship highlights its ability to detect nutritional-metabolic changes, a dimension potentially missed by NLR/PLR's sole focus on leukocyte activity. Our study is the first to focus on individuals with diabetes/prediabetes and demonstrate an L-shaped relationship between the HALP score and mortality. While these nonlinear analyses were exploratory, they were motivated by prior biological evidence and provide actionable thresholds for risk stratification. This study included 19,350 United States adults with diabetes/prediabetes (NHANES 2005–2018; median 84.2-month follow-up). Analyses revealed L-shaped relationships between the HALP score and all-cause/cardiovascular mortality. Multivariable-adjusted HRs increased sharply when HALP fell below thresholds of 42.29 (all-cause) and 39.98 (cardiovascular). Consistently, prior studies have demonstrated an L-shaped relationship between the HALP score and adverse outcomes. Hong Pan et al. observed an L-shaped association between the HALP score and cardiovascular mortality in the general population, with a corresponding V-shaped pattern for all-cause mortality. These findings partially align with our cohort's results(Pan and Lin, 2023). Jixin Fu et al. reported an L-shaped relationship between the HALP score and both all-cause mortality and CVD mortality in cancer survivors, consistent with our findings(Fu et al., 2024). The mechanisms underlying the L-shaped relationship remain uncertain, but may include the following.

As a composite score, the HALP integrates levels of hemoglobin, albumin, lymphocytes, and platelets, reflecting a patient's nutritional status and inflammatory status. These factors are frequently altered in individuals with diabetes, thereby increasing the risk of CVD. Specifically, low hemoglobin levels reduce the oxygen-carrying capacity of the blood, leading to chronic hypoxia. Hypoxia-inducible factor-1α(HIF-1α) is stable in hypoxia and keeps activating glycolysis genes like Lactate Dehydrogenase A and Phosphoglycerate Kinase 1. But long-term over-activation causes lactic acid buildup and mitochondrial dysfunction, worsening the myocardial energy crisis(Zhao et al., 2023). HIF-1α also promotes vascular endothelial cell damage and atherosclerotic plaque formation by inducing reactive oxygen species production and the release of pro-inflammatory factors such as tumor necrosis factor alpha (TNF-α) and interleukin-6 (IL-6)(Zhao et al., 2023). These synergistic effects boost cardiovascular risks and mortality in diabetic populations(Arkew et al., 2023; Thomas et al., 2006). Lower serum albumin levels impair the body's capacity to bind and scavenge inflammatory cytokines (TNF-α, IL-6) and free radicals, thereby promoting chronic low-grade inflammation(Cabrerizo et al., 2015). Patients with diabetes typically exhibit chronic low-grade inflammation, characterized by activation of inflammatory cells (e.g., macrophages and neutrophils) and elevated levels of inflammatory factors such as TNF-αand IL-6(Lontchi-Yimagou et al., 2013). Lymphocyte deficiency can result in insufficient suppression of these inflammatory signals, thereby perpetuating a chronic low-grade inflammatory state. Chronic low-grade inflammation triggers endothelial dysfunction, oxidized low-density lipoprotein retention, macrophage polarization, and Matrix Metalloproteinase-driven fibrous cap thinning, escalating atherosclerotic plaque vulnerability and thrombosis risk(Danesh et al., 2000; Henein et al., 2022; Ronit et al., 2020). In addition, lymphocytes influence endothelial cell function by secreting cytokines and chemokines. Low lymphocyte levels may impair endothelial cell responsiveness to inflammatory signals, thereby triggering endothelial dysfunction and promoting vasoconstriction, thrombosis, and atherosclerosis(Biondi-Zoccai et al., 2003). Increased platelet activation and aggregation are prevalent in diabetic patients, thereby promoting thrombosis, exacerbating endothelial dysfunction, and accelerating atherosclerosis(Chen et al., 2024b; Kaur et al., 2018). In addition, diabetes is frequently linked to microvascular pathology, affecting haemodynamics and increasing the risk of microvascular occlusion, thereby heightening CVD risk(Geng et al., 2023).

This study has several strengths. First, we used a large sample size with a long follow-up period for multiple sensitive analyses with stable and reliable results. Secondly, the HALP score, a composite inflammatory-nutritional index, provides a holistic assessment superior to individual biomarkers, revealing critical associations between systemic inflammation, nutritional deficits, and mortality risk in diabetic or prediabetic populations. Again, our study has some limitations. First, residual confounding (e.g., unmeasured drug use) may affect the association between the HALP score and mortality, despite adjusting for known confounders. Baseline medication data might not reflect longitudinal changes during follow-up, introducing potential time-dependent confounding. Future studies with longitudinal pharmacological data are needed to clarify relationships between the HALP score, medication use, and mortality. Second, this is a study of patients with diabetes/prediabetes in the United States population, and whether it is generalisable to other populations requires prospective studies to further validate the association and underlying mechanisms.

5. Conclusions

In conclusion, the HALP score is a valuable predictor of the risk of all-cause and cardiovascular mortality in patients with diabetes/prediabetes. The HALP score showed an L-shaped association with both all-cause and cardiovascular mortality, and the inflection points for the HALP score and poor prognosis were 42.29 and 39.98, respectively. Prospective studies should be conducted to assess the predictive value of the HALP score and to investigate the mechanisms underlying the L-shaped relationship.

CRediT authorship contribution statement

Taotao Zhang: Writing – review & editing, Writing – original draft, Methodology, Formal analysis, Data curation. Peiqian Liu: Writing – review & editing, Supervision, Methodology, Conceptualization.

Ethical approval and consent to participate

The United States National Health and Nutrition Examination Survey (NHANES) protocol received approval from both the NHANES Institutional Review Board (IRB) and the National Center for Health Statistics (NCHS) Research Ethics Review Board. Written informed consent was obtained from all participants. As this secondary analysis utilized publicly accessible de-identified data, additional IRB approval was not required.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Footnotes

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.pmedr.2025.103101.

Appendix A. Supplementary data

Supplementary material

mmc1.docx (20.7KB, docx)

Data availability

The datasets used and analysed during the current study are publicly available from the National Health and Nutrition Examination Survey (NHANES) repository, https://wwwn.cdc.gov/nchs/nhanes/.

References

  1. Arkew M., Asmerom H., Gemechu K., Tesfa T. Global prevalence of Anemia among type 2 diabetic adult patients: a systematic review and Meta-analysis. Diabetes Metab Syndr Obes. 2023;16:2243–2254. doi: 10.2147/dmso.S421851. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bergman M., Abdul-Ghani M., Chan J., Schmidt M.I., Ha J., Kim S.S., Sherman A.S., Jagannathan R., Tuomilehto J. Staging schema for early diagnosis of prediabetes. Lancet Diabetes Endocrinol. 2024;12:873–876. doi: 10.1016/s2213-8587(24)00320-6. [DOI] [PubMed] [Google Scholar]
  3. Biondi-Zoccai G.G., Abbate A., Liuzzo G., Biasucci L.M. Atherothrombosis, inflammation, and diabetes. J. Am. Coll. Cardiol. 2003;41:1071–1077. doi: 10.1016/s0735-1097(03)00088-3. [DOI] [PubMed] [Google Scholar]
  4. Cabrerizo S., Cuadras D., Gomez-Busto F., Artaza-Artabe I., Marín-Ciancas F., Malafarina V. Serum albumin and health in older people: review and meta analysis. Maturitas. 2015;81:17–27. doi: 10.1016/j.maturitas.2015.02.009. [DOI] [PubMed] [Google Scholar]
  5. Chen X.L., Xue L., Wang W., Chen H.N., Zhang W.H., Liu K., Chen X.Z., Yang K., Zhang B., Chen Z.X., Chen J.P., Zhou Z.G., Hu J.K. Prognostic significance of the combination of preoperative hemoglobin, albumin, lymphocyte and platelet in patients with gastric carcinoma: a retrospective cohort study. Oncotarget. 2015;6:41370–41382. doi: 10.18632/oncotarget.5629. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chen D., Chen J., Zhou Q., Mi H., Liu G. Association of the hemoglobin, albumin, lymphocyte, and platelet score with the risk of erectile dysfunction: a cross-sectional study. Sci. Rep. 2024;14:15869. doi: 10.1038/s41598-024-66667-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chen P., Zhu M., Guo M., Shi D., Chen Z., Du J. Platelet to high density lipoprotein cholesterol ratio is associated with diabetes and prediabetes in NHANES 2005 to 2018. Sci. Rep. 2024;14:30082. doi: 10.1038/s41598-024-81637-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Danesh J., Whincup P., Walker M., Lennon L., Thomson A., Appleby P., Gallimore J.R., Pepys M.B. Low grade inflammation and coronary heart disease: prospective study and updated meta-analyses. Bmj. 2000;321:199–204. doi: 10.1136/bmj.321.7255.199. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Diagnosis and Classification of Diabetes Standards of Care in Diabetes-2025. Diabetes Care. 2025;48:S27–s49. doi: 10.2337/dc25-S002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Ding R., Zeng Y., Wei Z., He Z., Jiang Z., Yu J., You C. The L-shape relationship between hemoglobin, albumin, lymphocyte, platelet score and the risk of diabetic retinopathy in the US population. Front Endocrinol (Lausanne) 2024;15:1356929. doi: 10.3389/fendo.2024.1356929. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dong G., Gan M., Xu S., Xie Y., Zhou M., Wu L. The neutrophil-lymphocyte ratio as a risk factor for all-cause and cardiovascular mortality among individuals with diabetes: evidence from the NHANES 2003-2016. Cardiovasc. Diabetol. 2023;22:267. doi: 10.1186/s12933-023-01998-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fu J., Yue X., Zou Y., Zhang J., Wang X., Zhang D. Association of hemoglobin, albumin, lymphocyte, and platelet score with risk of all-cause and cause-specific mortality among cancer survivors: NHANES 1999-2018. Front. Oncol. 2024;14:1402217. doi: 10.3389/fonc.2024.1402217. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Geng T., Zhu K., Lu Q., Wan Z., Chen X., Liu L., Pan A., Liu G. Healthy lifestyle behaviors, mediating biomarkers, and risk of microvascular complications among individuals with type 2 diabetes: a cohort study. PLoS Med. 2023;20 doi: 10.1371/journal.pmed.1004135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Henein M.Y., Vancheri S., Longo G., Vancheri F. The role of inflammation in cardiovascular disease. Int. J. Mol. Sci. 2022;23 doi: 10.3390/ijms232112906. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Kaur R., Kaur M., Singh J. Endothelial dysfunction and platelet hyperactivity in type 2 diabetes mellitus: molecular insights and therapeutic strategies. Cardiovasc. Diabetol. 2018;17:121. doi: 10.1186/s12933-018-0763-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Lan N.S.R., Dwivedi G., Fegan P.G., Game F., Hamilton E.J. Unravelling the cardio-renal-metabolic-foot connection in people with diabetes-related foot ulceration: a narrative review. Cardiovasc. Diabetol. 2024;23:437. doi: 10.1186/s12933-024-02527-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lee G., Choi S., Kim K., Yun J.M., Son J.S., Jeong S.M., Kim S.M., Park S.M. Association of Hemoglobin Concentration and its Change with Cardiovascular and all-cause mortality. J. Am. Heart Assoc. 2018;7 doi: 10.1161/jaha.117.007723. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Li H., Li Y., Guo W., Liu X., Wang Y., Zeng T., Kong W. Monocyte-lymphocyte ratio predicts cardiovascular diseases death in individuals with type 2 diabetes. J Diabetes Investig. 2024 doi: 10.1111/jdi.14329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Li W., Yan Y., Cui X., Bian J., Yuan L., Wang G. Exploring the association between hemoglobin, albumin, lymphocyte, and platelet score and all-cause mortality among middle-aged and older patients with osteoarthritis. J. Invest. Med. 2025;73:94–103. doi: 10.1177/10815589241273682. [DOI] [PubMed] [Google Scholar]
  20. Liu C., Liang D. The association between the triglyceride-glucose index and the risk of cardiovascular disease in US population aged ≤ 65 years with prediabetes or diabetes: a population-based study. Cardiovasc. Diabetol. 2024;23:168. doi: 10.1186/s12933-024-02261-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Liu L., Gong B., Wang W., Xu K., Wang K., Song G. Association between haemoglobin, albumin, lymphocytes, and platelets and mortality in patients with heart failure. ESC Heart Fail. 2024;11:1051–1060. doi: 10.1002/ehf2.14662. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lontchi-Yimagou E., Sobngwi E., Matsha T.E., Kengne A.P. Diabetes mellitus and inflammation. Curr. Diab. Rep. 2013;13:435–444. doi: 10.1007/s11892-013-0375-y. [DOI] [PubMed] [Google Scholar]
  23. Marassi M., Fadini G.P. The cardio-renal-metabolic connection: a review of the evidence. Cardiovasc. Diabetol. 2023;22:195. doi: 10.1186/s12933-023-01937-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Mathur K., Kurbanova N., Qayyum R. Platelet-lymphocyte ratio (PLR) and all-cause mortality in general population: insights from national health and nutrition education survey. Platelets. 2019;30:1036–1041. doi: 10.1080/09537104.2019.1571188. [DOI] [PubMed] [Google Scholar]
  25. Matoori S. Diabetes and its complications. ACS Pharmacol Transl Sci. 2022;5:513–515. doi: 10.1021/acsptsci.2c00122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Pan H., Lin S. Association of hemoglobin, albumin, lymphocyte, and platelet score with risk of cerebrovascular, cardiovascular, and all-cause mortality in the general population: results from the NHANES 1999-2018. Front Endocrinol (Lausanne) 2023;14:1173399. doi: 10.3389/fendo.2023.1173399. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Raghavan S., Vassy J.L., Ho Y.L., Song R.J., Gagnon D.R., Cho K., Wilson P.W.F., Phillips L.S. Diabetes mellitus-related all-cause and cardiovascular mortality in a National Cohort of adults. J. Am. Heart Assoc. 2019;8 doi: 10.1161/jaha.118.011295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Ronit A., Kirkegaard-Klitbo D.M., Dohlmann T.L., Lundgren J., Sabin C.A., Phillips A.N., Nordestgaard B.G., Afzal S. Plasma albumin and incident cardiovascular disease: results from the CGPS and an updated Meta-analysis. Arterioscler. Thromb. Vasc. Biol. 2020;40:473–482. doi: 10.1161/atvbaha.119.313681. [DOI] [PubMed] [Google Scholar]
  29. Seidu S., Kunutsor S.K., Khunti K. Serum albumin, cardiometabolic and other adverse outcomes: systematic review and meta-analyses of 48 published observational cohort studies involving 1,492,237 participants. Scand. Cardiovasc. J. 2020;54:280–293. doi: 10.1080/14017431.2020.1762918. [DOI] [PubMed] [Google Scholar]
  30. Sun H., Saeedi P., Karuranga S., Pinkepank M., Ogurtsova K., Duncan B.B., Stein C., Basit A., Chan J.C.N., Mbanya J.C., Pavkov M.E., Ramachandaran A., Wild S.H., James S., Herman W.H., Zhang P., Bommer C., Kuo S., Boyko E.J., Magliano D.J. IDF diabetes atlas: global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Res. Clin. Pract. 2022;183 doi: 10.1016/j.diabres.2021.109119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Thomas M.C., Cooper M.E., Rossing K., Parving H.H. Anaemia in diabetes: is there a rationale to TREAT? Diabetologia. 2006;49:1151–1157. doi: 10.1007/s00125-006-0215-6. [DOI] [PubMed] [Google Scholar]
  32. Won K.B., Choi S.Y., Chun E.J., Park S.H., Sung J., Jung H.O., Chang H.J. Longitudinal assessment of serum albumin levels with the risk of coronary artery calcification progression in an asymptomatic population of Korean adults: an observational cohort study. BMJ Open. 2024;14 doi: 10.1136/bmjopen-2024-086075. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Xu H., Zheng X., Ai J., Yang L. Hemoglobin, albumin, lymphocyte, and platelet (HALP) score and cancer prognosis: a systematic review and meta-analysis of 13,110 patients. Int. Immunopharmacol. 2023;114 doi: 10.1016/j.intimp.2022.109496. [DOI] [PubMed] [Google Scholar]
  34. Xu M., Chen L., Hu Y., Wu J., Wu Z., Yang S., Kang W., He J., Ren W. The HALP (hemoglobin, albumin, lymphocyte, and platelet) score is associated with early-onset post-stroke cognitive impairment. Neurol. Sci. 2023;44:237–245. doi: 10.1007/s10072-022-06414-z. [DOI] [PubMed] [Google Scholar]
  35. Yang C., Yang Q., Xie Z., Peng X., Liu H., Xie C. Association of systemic immune-inflammation-index with all-cause and cause-specific mortality among type 2 diabetes: a cohort study base on population. Endocrine. 2024;84:399–411. doi: 10.1007/s12020-023-03587-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Yu B., Li M., Yu Z., Zheng T., Feng X., Gao A., Zhang H., Gao R. The non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio (NHHR) as a predictor of all-cause and cardiovascular mortality in US adults with diabetes or prediabetes: NHANES 1999-2018. BMC Med. 2024;22:317. doi: 10.1186/s12916-024-03536-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Zhang D., Chen S., Cao W., Geng N., Feng C. HALP score based on hemoglobin, albumin, lymphocyte and platelet can predict the prognosis of tongue squamous cell carcinoma patients. Heliyon. 2023;9 doi: 10.1016/j.heliyon.2023.e20126. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Zhang T., Liu W., Xu C. Correlation analysis of hemoglobin, albumin, lymphocyte, platelet score and platelet to albumin ratio and prognosis in patients with lung adenosquamous carcinoma. Front. Oncol. 2023;13:1166802. doi: 10.3389/fonc.2023.1166802. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Zhao Y., Xiong W., Li C., Zhao R., Lu H., Song S., Zhou Y., Hu Y., Shi B., Ge J. Hypoxia-induced signaling in the cardiovascular system: pathogenesis and therapeutic targets. Signal Transduct. Target. Ther. 2023;8:431. doi: 10.1038/s41392-023-01652-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Zhao P., Du T., Zhou Q., Wang Y. Association of weight-adjusted-waist index with all-cause and cardiovascular mortality in individuals with diabetes or prediabetes: a cohort study from NHANES 2005-2018. Sci. Rep. 2024;14:24061. doi: 10.1038/s41598-024-74339-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zheng Y., Huang Y., Li H. Hemoglobin albumin lymphocyte and platelet score and all-cause mortality in coronary heart disease: a retrospective cohort study of NHANES database. Front Cardiovasc Med. 2023;10:1241217. doi: 10.3389/fcvm.2023.1241217. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Zhou J., Yang D. Prognostic significance of hemoglobin, albumin, lymphocyte and platelet (HALP) score in hepatocellular carcinoma. J Hepatocell Carcinoma. 2023;10:821–831. doi: 10.2147/jhc.S411521. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Zhou X., Hou G., Wang X., Peng Z., Yin X., Yang J., Wang S., He Y., Wang Y., Sui J., Qiang W., Guo H., Wang Y., Lin L., Shi B., He M. Metabolomic studies reveal and validate potential biomarkers of diabetic retinopathy in two Chinese datasets with type 2 diabetes: a cross-sectional study. Cardiovasc. Diabetol. 2024;23:439. doi: 10.1186/s12933-024-02535-1. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary material

mmc1.docx (20.7KB, docx)

Data Availability Statement

The datasets used and analysed during the current study are publicly available from the National Health and Nutrition Examination Survey (NHANES) repository, https://wwwn.cdc.gov/nchs/nhanes/.

Articles from Preventive Medicine Reports are provided here courtesy of Elsevier

RESOURCES