Abstract
The subjective indicator of health self-rated health (SRH) and the chronic inflammation biomarker soluble urokinase plasminogen activator receptor (suPAR) are both robust predictors of healthcare use and mortality. However, the possible relationship between SRH and suPAR in the assessment of hospitalization and mortality risk is unknown. We used data from the Danish population-based Inter99 cohort to examine the association between SRH and suPAR and test their individual and combined associations with 2-year risk of acute hospitalization and 5- and 15-year mortality. SRH and serum suPAR levels were measured in 5490 participants (median age 45.1 years, 48.7% men). Poorer SRH was associated with elevated suPAR. In unadjusted analyses, SRH and suPAR were individually associated with higher risks of acute hospitalization and mortality, and both measures remained independently associated with higher risks of hospitalization and 15-year mortality after mutual adjustments. The association of suPAR with mortality was stronger in poorer SRH categories, and when combined, SRH and suPAR could identify different groups of individuals with increased risk of acute hospitalization and mortality. Both SRH and suPAR were independently associated with risk of acute hospitalization and mortality, and different combinations of the two measures could identify different groups of individuals at increased risk.
Similar content being viewed by others
Introduction
There is solid evidence that a person’s self-evaluation of their overall health status is a strong predictor of morbidity, use of health services, and mortality independently of objective measures of health and other risk factors such as socioeconomic status, age, chronic illness, gender, and physical examination1,2,3,4,5,6. Self-rated health (SRH) is a simple subjective measure capturing a condensed summary of experienced somatic information, and is thought to have a biological foundation, however, these processes are not yet well identified7.
Studies have suggested that “sickness signals” such as changes in biomarker levels influence how the body or brain interpret general health, and that a deviation from normal values might negatively impact a person’s SRH8,9,10. In a recent study, researchers found a significant association between SRH and 57 out of 150 investigated blood and urine biomarkers encompassing a wide range of biological domains, and 26 of the biomarkers (including the inflammation marker C-reactive protein [CRP]) remained significantly associated with SRH after adjusting for chronic diseases and physical functioning10. Several other studies have reported associations between SRH and biomarkers of systemic inflammation, including cytokines8,11,12,13, CRP10,11,12, white blood cell counts9. A small number of studies further examined whether biomarker levels explained the strong association between SRH and mortality, and all reported that after adjusting for biomarkers, the association was only slightly weakened and SRH remained independently associated with mortality9,10,14,15,16. Only two of these studies have compared the performance of SRH and biomarkers in the prediction of mortality and found that while SRH was a better predictor of mortality than biomarkers14,15, the combination of SRH with biomarkers improved the predictive ability of the models14.
The soluble form of the urokinase plasminogen activator receptor (suPAR) is a novel biomarker of systemic chronic inflammation17, and a strong, unspecific, prognostic biomarker for morbidity and mortality18,19. suPAR levels are elevated with many chronic diseases18,20,21, and shown to positively correlate with other inflammatory biomarkers and remain a strong predictor of adverse health outcomes even after adjustments for inflammatory biomarkers such as IL-6 and CRP18,22,23,24,25,26. suPAR levels are elevated by unhealthy lifestyle such as smoking and an unhealthy diet19. However, change of lifestyle habits towards a healthier lifestyle results in lowering of suPAR, and the resultant suPAR is predictive of mortality19,27. Furthermore, a recent study demonstrated that suPAR is a causal factor in the development of cardiovascular disease28. This indicates that suPAR is an early warning biomarker and that interventions modifying suPAR can lower the risk of adverse outcomes.
Both SRH and suPAR are unspecific measures of general health and strong prognostic markers of outcome. Poor SRH is correlated with higher suPAR24,29, but the relationship between the two biomarkers has not been thoroughly investigated, and it is unclear whether they reflect common or independent processes related to increased risk of illness and mortality. The present study aimed to (i) examine the association between SRH and suPAR, (ii) test whether SRH and suPAR are independently associated with the risk of acute hospitalization and mortality, as well as (iii) test the interaction between SRH and suPAR in their association with the risk of acute hospitalization and mortality.
Results
Participant characteristics
Of the 6784 individuals who participated in the baseline examination, 6717 had answered the SRH question, and 5540 had a serum suPAR measurement available. A total of 5490 individuals had data available for both SRH and suPAR levels and were included in this study (Fig. 1). Baseline characteristics of the participants according to the three categories of SRH or suPAR are presented in Table 1. The characteristics of the participants according to the four combinations of good/poor SRH and high/low suPAR are presented in Supplementary Table 1.
Relationship between SRH and suPAR levels
The participants had a median suPAR level of 3.4 ng/mL (IQR: 2.7–4.3), and suPAR levels increased with worse SRH (Fig. 2). When reducing SRH into 3 categories, suPAR levels for participants reporting excellent/very good SRH were 3.2 ng/mL (IQR: 2.6–4.0), 3.4 ng/mL (IQR: 2.8–4.3) for those reporting good SRH, and 3.8 ng/mL (IQR: 3.1–4.9) for those reporting fair/bad SRH (Supplementary Table 2). The number of participants within each of the SRH categories according to their suPAR category is also shown in Supplementary Table 2.
Individual and mutually adjusted associations of SRH and suPAR with the risk of acute hospitalization
Within 2 years after baseline examination, 437 (8.0%) participants were acutely hospitalized, 8 (0.2%) had died, and 10 (0.2%) were lost to follow-up. When suPAR levels were analyzed as a continuous variable, a doubling in suPAR levels was associated with a significantly higher risk of acute hospitalization both in unadjusted (Sub-distribution HR: 1.50, 95% CI: 1.27 to 1.78, P < 0.0001) and adjusted (SDHR: 1.30, 95% CI: 1.06 to 1.59, P = 0.011) analyses. The interaction between suPAR and sex was not statistically significant (P = 0.46). In individual analyses of SRH and suPAR as 3-level categories, participants reporting good or fair/bad SRH, as well as those with intermediate or high suPAR, were at significantly higher risk of hospitalization compared to participants with excellent/very good SRH or low suPAR, respectively (Fig. 3A). Mutual adjustment for SRH and suPAR did not affect the strength of the associations (Fig. 3A). To further investigate whether SRH and suPAR were independently associated with the risk of acute hospitalization, we added an interaction term between suPAR and SRH categories to the mutually adjusted model. The interaction term was not statistically significant, suggesting that the associations of suPAR or SRH categories with the risk of acute hospitalization are not significantly dependent on each other (P = 0.62).
Combined associations of SRH and suPAR with the risk of acute hospitalization
We also studied the combined associations of SRH and suPAR with the risk of acute hospitalization by stratifying the participants into four or nine groups based on good or poor SRH and high or low suPAR. For participants reporting good SRH, having high suPAR levels did not increase the risk of hospitalization compared to having low suPAR (Fig. 3B). However, poor SRH increased the risk of acute hospitalization for participants with low suPAR, but also for those with high suPAR. The combination of poor SRH and high suPAR presented the highest sub-distribution hazard ratios for hospitalization risk, although CIs overlapped with other combinations (Fig. 3B). Analyses according to the nine combinations of SRH and suPAR groups are presented in Supplementary Table 3.
Individual and mutually adjusted associations of SRH and suPAR with mortality
Five years after the baseline examination, 60 (1.1%) participants had died and 19 (0.4%) were lost to follow-up. A doubling in suPAR levels was associated with a significantly higher risk of 5-year mortality in unadjusted (HR: 2.03, 95% CI: 1.24 to 3.33, P = 0.005) analyses, but not in analyses adjusted for age, sex, smoking status, BMI, and comorbidities (HR: 1.47, 95% CI: 0.78 to 2.76, P = 0.23). The interaction between suPAR levels and sex was not statistically significant (P = 0.79). In unadjusted analysis of SRH categories, participants reporting fair/bad SRH, but not those reporting good SRH, had a significantly higher risk of 5-year mortality compared to participants with excellent/very good SRH (Fig. 4A). For suPAR categories, participants with intermediate and high suPAR had a significantly higher risk of 5- year mortality compared to those with low suPAR (Fig. 4A). After controlling for age, sex, smoking status, BMI, and comorbidities as well as in the mutually adjusted model, the associations no longer reached statistical significance (Fig. 4A). In the mutually adjusted model, the interaction terms for suPAR and SRH categories was statistically significant (P < 0.0001; Fig. 5A).
Fifteen years after the baseline examination, 326 (5.9%) participants had died and 47 (0.9%) were lost to follow-up. A doubling in suPAR levels was significantly associated with a higher risk of 15-year mortality both in unadjusted (HR: 2.32, 95% CI: 1.89 to 2.85, P < 0.0001) and adjusted (HR: 1.81, 95% CI: 1.40 to 2.33, P < 0.0001) analyses. When the interaction term between suPAR levels and sex was included (P = 0.03), higher suPAR levels were significantly associated with a higher risk of 15-year mortality for men (HR: 2.17, 95% CI: 1.60 to 2.95, P < 0.0001), but not for women (HR: 1.29, 95% CI: 0.86 to 1.92, P = 0.22). In univariate analyses of SRH or suPAR categories, participants with good and fair/bad SRH as well as those with intermediate and high suPAR had a significantly higher risk of 15-year mortality compared to participants with excellent/very good SRH or low suPAR, respectively (Fig. 4B). The associations remained statistically significant after adjustments for age, sex, smoking, BMI, and comorbidities and mutual adjustments for suPAR and SRH only slightly weakened the associations (Fig. 4B). The interaction terms for suPAR and SRH categories was statistically significant (P = 0.02). Figure 5B shows the HRs for each suPAR category within each SRH category, with low suPAR in the excellent/very good SRH category as the reference. Increasing suPAR levels had a larger effect on the risk of 15-year mortality for participants reporting good and fair/bad SRH than for those reporting excellent/very good SRH (Fig. 5B).
Combined associations of SRH and suPAR with mortality
When stratifying the participants according to the four combinations of SRH and suPAR groups, only those with both poor SRH and high suPAR had a significantly higher risk of 5-year mortality compared to those with good SRH and low suPAR (Fig. 4C). For 15-year mortality, compared to participants with both good SRH and low suPAR, participants with either poor SRH or high suPAR had a higher risk of 15-year mortality, and the combination of both poor SRH and high suPAR presented the highest hazard ratios, although CIs overlapped with other combinations (Fig. 4D). Analyses according to the nine combinations of SRH and suPAR groups are presented in Supplementary Table 3.
Discussion
In this large population-based study, we tested the individual and combined associations of a subjective indicator (SRH) and an objective biomarker (suPAR) of health with the risk of acute hospitalization, short- and long-term mortality. The main finding of our study is that both SRH and suPAR were strongly and independently associated with 2-year risk of acute hospitalization and long-term mortality after adjustments for age, sex, smoking, BMI, and comorbidities, and that the strength of these associations was unchanged by mutual adjustments for suPAR and SRH, respectively. Furthermore, we found that overall, different combinations of poorer SRH and/or elevated suPAR were associated with higher risks of acute hospitalization and mortality compared to the combination of the best category of SRH and the lowest suPAR levels.
Similar to findings with other inflammation biomarkers8,9,10,11,12,13,30,31,32, and previous reports on suPAR24,29, we found that suPAR levels increased with poorer SRH. Kananen et al. proposed three pathways by which biomarkers could influence SRH10. First, individuals may be aware of their levels of certain biomarkers related to specific diagnoses such as cholesterol or glucose levels and may take these biomarkers-or the associated diagnoses-into account in their assessment of health10. Although participants would not have been aware of their suPAR levels at the time of SRH assessment, it is possible that they would take into account other diagnoses or biomarker levels that are known to be associated with elevated suPAR, such as diabetes, cardiovascular disease, elevated cholesterol18,33,34,35. Second, individuals may consider lifestyle and behavioral factors in their health assessment10. We have not tested whether lifestyle factors influenced the relationship between suPAR and SRH. But this is a possibility as suPAR levels are known to be dependent on lifestyle36 and can be affected by lifestyle changes19. Third, changes in biomarker levels can influence the perception of health, and in particular, changes in biomarkers of inflammation including cytokines are thought to be involved in symptoms of sickness such as fatigue, loss of appetite, and weakness10,37,38,39. This mechanism has been suggested to explain part of the relationship between SRH and mortality10.
The individual relationships of poor SRH and elevated suPAR levels with risk of acute hospitalization have been described5,6,40,41,42, but the two measures have not previously been compared or combined. In the present study, poor SRH and high suPAR were associated with a higher risk of acute hospitalization independently of each other, though the effect of poor SRH was greater than that of high suPAR. Furthermore, poor SRH was significantly associated with a higher risk of acute hospitalization when combined with either high or low suPAR, whereas high suPAR did not increase risk for individuals with good SRH. Although both SRH and suPAR reflect disease processes, the stronger association of SRH with hospitalization risk may not be surprising as individuals faring badly, for example perceiving a greater illness burden and impaired function, may be more likely to actively seek care and be acutely hospitalized, while elevated chronic inflammation might more strongly reflect slowly developing pathological processes leading to increased biological aging and long-term adverse outcomes. In older adults, a composite score of inflammation (based on albumin, cholesterol, CRP, and IL-6) was a better predictor of mortality than of hospitalization over a 4-year follow-up43. A large survey investigating the association of SRH with healthcare use concluded that the feeling of poor health was more strongly associated with risk of hospitalization than diagnosed chronic conditions or age44. Thus, the choice to actively seek medical care may be more guided by interoceptive signals or worries, and other behaviors, that are not directly informed by objective health indicators, including chronic inflammation. Nevertheless, poor SRH was also strongly associated with 5- and 15-year mortality, indicating that the feeling of poor health is also grounded in real risk of mortality.
We observed strong associations of poor SRH and high suPAR with mortality, which have also been repeatedly reported in previous studies2,4,18. A small number of studies investigated whether the SRH-mortality association was explained by biomarkers such as CRP or panels that included CRP; in all these studies, adjustment for biomarkers only slightly weakened the associations, and SRH remained statistically associated with mortality10,14,15,16. In the present study, the HRs for SRH and mortality risk were largely unaffected when controlling for suPAR. Similarly, the relationship of suPAR with mortality was unaffected after adjusting for SRH. Thus, our—and previous—findings suggest that the association of SRH with mortality cannot be explained by biomarkers, such as suPAR, alone; and the association between biomarkers such as suPAR and mortality cannot be explained by SRH alone7,15. Furthermore, in contrast to the reports that SRH is more strongly associated with mortality than biomarker panels14,15, in our study, the association of high suPAR with mortality was larger than that of fair/bad SRH. Recent evidence suggests that in contrast to traditional inflammation biomarkers such as CRP and cytokines that reflect acute inflammation, suPAR levels increase and decrease at a slower speed, are less affected by acute changes, and are thought to capture other aspects of chronic inflammation17. It is also still unknown whether suPAR is involved in sickness behavior to the same extent as cytokines and acute-phase proteins. Finally, we found that a combination of both suPAR and SRH provided greater information about the risk of mortality than either measure alone; with suPAR adding information about the risk of mortality for participants with excellent to good SRH, and likewise, poorer SRH could identify participants with low suPAR who were at a higher risk of mortality. Furthermore, higher suPAR and poor SRH appeared to impact the risk of 15-year mortality in a synergistic fashion-for participants with good to bad SRH, the risk of mortality further increased with increasing suPAR. Therefore, even though SRH and suPAR levels were related, in this study, the two measures appear to provide unique information regarding health status and may reflect independent pathways to mortality.
Despite being a validated research tool and simple to administer, SRH is not widely used in clinical settings. Incorporating assessment of patient-reported outcomes such as SRH in routine care may help identify at-risk individuals that may otherwise be overlooked by routine biochemical evaluation alone, and those whose complaints during examination are incorrectly interpreted by healthcare professionals. In our study, poor SRH alone, or high suPAR alone, identified different groups of participants, with different risks of hospitalization and mortality, and the combination of both poor SRH and suPAR was consistently associated with an increased risk for all outcomes. Furthermore, in interaction analyses, we found that suPAR was more strongly associated with mortality for participants reporting poor SRH. Thus, the combination of the two biomarkers provided synergistic information on the health status, compared to each measure alone. This may have clinical implications in future intervention studies in the general population, allowing for intervention in this particularly high-risk group of individuals. It has previously been shown that a lifestyle change in the Inter99 cohort, i.e., smoking cessation resulted in lowering of suPAR levels (from baseline to 5-year samples) and that the resultant 5-year suPAR value was associated with future mortality19. A recent clinical trial used elevated suPAR levels as a guide for identifying patients with increased risk of progression of COVID-19 and showed a decreased length of hospitalization and mortality for these patients after treatment with an interleukin-1 inhibitor45. This indicates that early identification and intervention of high-risk individuals can lead to lowering of risk of unplanned hospitalization and mortality. By combining suPAR and SRH data in the present study, we have significantly enhanced risk identification, possibly leading to a smaller, but more targeted, group of individuals that may be offered intervention (e.g., only those with poor SRH and high suPAR) in future general population intervention studies.
This study has limitations. First, despite a large sample size, this general population cohort was relatively young and healthy, and only 60 (1.1%) participants died during the first 5 years after data collection. Second, other biomarkers of inflammation, such as CRP and cytokines, could have been included in the study for comparison with suPAR as it has been suggested that suPAR and CRP or cytokines may reflect different aspects of inflammation17. Third, due to the cross-sectional design of this study, we could not determine whether there was a causal relationship between high suPAR levels and poorer SRH, nor the direction of this relationship, or whether the association between the two could be related to sickness behavior.
Our results show that individuals with poor SRH, high suPAR, or a combination of both poor SRH and high suPAR had increased risk for acute hospitalization and long-term mortality. These findings contribute to the body of knowledge promoting the combined use of subjective and objective indicators of health to assess the risk of healthcare use and mortality.
Methods
Study and participants
This study is based on data collected from participants from the Danish Inter99 cohort, a non-pharmacological intervention study for the prevention of ischemic heart disease consisting of 13,016 participants aged 30–60 years who were randomized to either a high or a low intervention of lifestyle counselling. Of these, 6784 agreed to participate and underwent a baseline examination between March 1999 and January 200146. The examination included a clinical examination, extensive questionnaires assessing lifestyle, socioeconomic status, mental health, and SRH, as well as blood sampling. Participants were linked to the Danish health and social registers containing information on hospitalizations, ICD-10 codes, and death46,47. The study was conducted in accordance with the Declaration of Helsinki and was approved by the Scientific Ethics Committee of the Capital Region of Denmark (KA 98 155) and the Danish Data Protection Agency, and was registered as a clinical trial (ClinicalTrials.gov; NCT00289237). All participants provided written informed consent before taking part in the study. No effect of the lifestyle intervention was observed in the original study47, and we included all participants who had data available for both SRH and suPAR levels in the present study (n = 5490, Fig. 1).
Self-rated health
SRH was assessed at baseline via a self-administered questionnaire. The question was formulated as: “How do you think your health is, all in all?”, with the following possible answers: “excellent”, “very good”, “good”, “fair”, or “bad”.
suPAR
Serum levels of suPAR (ng/mL) at baseline were measured using the suPARnostic ELISA assay (Virogates, Birkeroed, Denmark) as previously reported36. Samples were measured in singlets, but according to the manufacturer of the suPARnostic ELISA, the intra‐assay variation is 2.8%, and the inter-assay variation as 9.2%. Samples with values below the detectable range of the assay (0.6 ng/mL) were excluded (n = 13), and samples with values above the detectable range were assigned a value corresponding to the upper limit of quantification (22 ng/mL, n = 2).
Socioeconomic and lifestyle variables, and comorbidities
The participants’ socioeconomic status, smoking, alcohol consumption, and physical activity habits were assessed at baseline via self-administered questionnaires. The socioeconomic status of the participants was based on the duration of their education after primary school and categorized as low (< 2 years), medium (2–4 years), or high (> 4 years)47. Smoking was categorized as: never smoker, former smoker, occasional, or daily smoker36; alcohol consumption as abstinent (0 units per week), within recommendations (1–21 weekly units for men, or 1–14 for women), or overuse (> 21 weekly units for men, or > 14 units for women)36, according to recommendations on alcohol consumption at the time of assessment; and physical activity as low (mainly sedentary), light (walks, biking, or light activity for up to 4 h per week), moderate (moderate activity at least 3 times per week), or high (competitive sports or long-distance running several times per week)19. Body mass index (BMI, kg/m2) was calculated from weight and height measurements without shoes. Data on comorbidities included self-reported diagnoses of hypertension, diabetes and cardiovascular disease other than myocardial infarction. A variable for comorbidities including diagnoses of diabetes and cardiovascular disease was generated.
Outcome variables
Participants were followed from the time of baseline examination (March 1999 to January 2001) until 31 December 2016, for the outcomes of acute hospitalization within 2 years as well as 5- and 15-year mortality. Acute hospitalizations within 2 years following baseline examination were identified from the Danish National Patient Register (NPR) as “admission type” coded as acute admissions, and “patient type” coded as 24 h patient, and the time elapsed between the acute hospitalization dates and the date of baseline examination was calculated. If participants had several acute hospitalizations, the acute hospitalization date with the shortest time elapsed after the baseline examination date was selected. Hospitalizations for Injury, poisoning, and certain other consequences of external causes (ICD10 codes: S00-T98) as well as Pregnancy, childbirth, and the puerperium (ICD10 codes: O00-O99) were excluded. Survival status was obtained from the vital status and date of change in vital status variables from the Danish Civil Registration System (CPR). Survival time for participant with vital status coded as “dead” was calculated as the time elapsed between the baseline examination date and the date of death recorded in the register.
Statistical analysis
Descriptive data of participants are presented as the median and interquartile range (IQR) for continuous variables or the count and percentage for categorical variables.
We explored the relationship between SRH and suPAR using a one-way ANOVA for comparison of suPAR levels between the SRH category “excellent” and the other four categories. suPAR levels were log-transformed using log2, and back-transformed estimates are reported.
We used Cox regression analyses to assess the independent and combined associations of SRH and suPAR with the risk of acute hospitalization within 2-year after baseline, and of 5- and 15-year mortality. Loss to follow-up was considered a censored event, and for the analysis of acute hospitalization risk, death was considered a competing event. For the analyses of acute hospitalization risk, we report sub-distribution hazard ratio (SDHR) and 95% confidence intervals (CI) obtained using the Fine and Gray competing risk regression models. For mortality risk analyses, we report hazard ratios (HRs) with 95% confidence intervals (CIs).
To assess the independent associations of SRH and suPAR with the outcomes, suPAR was log2-transformed for analysis as a continuous variable and tested in unadjusted models and models adjusted for age, sex, smoking, BMI, and comorbidities; the interaction between suPAR levels and sex was also tested in the adjusted models. The SRH measure was reduced to 3 categories: excellent/very good (n = 1830, 33.3%), good (n = 3087, 56.2%), and fair/bad (n = 573, 10.4%). Clinical cut-offs for suPAR have not yet been established. Therefore, and to ease comparison between SRH and suPAR, we created a categorical variable for suPAR where participants were stratified into three categories of the same size as the three SRH categories: low suPAR (range: 0.649–2.950 ng/mL, n = 1830, 33.3%), intermediate suPAR (> 2.950–< 5.460 ng/mL, n = 3087, 56.2%), and high suPAR (5.460–22 ng/mL, n = 573, 10.4%). The associations of 3-level SRH and suPAR were tested individually in unadjusted models and models adjusted for age, sex, smoking, BMI and comorbidities, as well as in a mutually adjusted model that included both the 3-level SRH and suPAR, age, sex, smoking, BMI and comorbidities. In addition, we tested the associations of suPAR categories with the outcomes within each of the SRH categories by adding an interaction term for SRH and suPAR in the mutually adjusted models.
Finally, to assess the individual and combined associations of SRH and suPAR with the outcomes, we stratified the participants into four groups: (i) good SRH and low suPAR, n = 4434 (80.8%), (ii) good SRH and high suPAR, n = 483 (8.8%); (iii) poor SRH and low suPAR, n = 483 (8.8%); and (iv) poor SRH and high suPAR, n = 90 (1.6%). The good SRH group consisted of participants who answered “excellent”, “very good”, or “good” (n = 4917, 89.6%), and the poor SRH group consisted of participants who answered “fair” or “bad” (n = 573, 10.4%) to the questionnaire. The low suPAR group consisted of participants from the previously defined low and intermediate suPAR categories (n = 4917, 86.6%), and the high suPAR group consisted of participants in the previously defined high suPAR category (n = 573, 10.4%). Furthermore, we stratified the participants into nine groups (based on the 3 levels of SRH and the 3 levels of suPAR). We tested the associations of the four and the nine combined SRH and suPAR groups with the outcomes in unadjusted and adjusted (age, sex, smoking, BMI, and comorbidities) models.
For statistical analysis, we used SAS Enterprise Guide version 7.15 (SAS Institute, Cary, NC, USA). Graphs were made in GraphPad Prism version 9 (GraphPad Software Inc., San Diego, CA, USA). Two-sided p-values are reported, and statistical significance was defined as a p-value < 0.05.
Data availability
The datasets are not publicly available due to regulations set out by the Danish Data Protection Agency but are available from J.P. on reasonable request.
References
Desalvo, K. B., Bloser, N., Reynolds, K., He, J. & Muntner, P. Mortality prediction with a single general self-rated health question. J. Gen. Intern. Med. 20, 267–275 (2005).
Idler, E. L. & Benyamini, Y. Self-rated health and mortality: A review of twenty-seven community studies. J. Health Soc. Behav. 38, 21–37 (1997).
Benyamini, Y. & Idler, E. L. Community studies reporting association between self-rated health and mortality: Additional studies, 1995 to 1998. Res. Age. 21, 392–401 (1999).
DeSalvo, K. B., Fan, V. S., McDonell, M. B. & Fihn, S. D. Predicting mortality and healthcare utilization with a single question. Health Serv. Res. 40, 1234–1246 (2005).
Kennedy, B. S., Kasl, S. V. & Vaccarino, V. Repeated hospitalizations and self-rated health among the elderly: A multivariate failure time analysis. Am. J. Epidemiol. 153, 232–241 (2001).
Tamayo-Fonseca, N. et al. Self-rated health and hospital services use in the Spanish National Health System: A longitudinal study. BMC Health Serv. Res. 15, 492 (2015).
Jylhä, M. Social science and medicine what is self-rated health and why does it predict mortality? Towards a unified conceptual model q. Soc. Sci. Med. 69, 307–316 (2009).
Lekander, M., Elofsson, S., Neve, I.-M., Hansson, L.-O. & Undén, A.-L. Self-rated health is related to levels of circulating cytokines. Psychosom Med. 66, 559–563 (2004).
Jylhä, M., Volpato, S. & Guralnik, J. M. Self-rated health showed a graded association with frequently used biomarkers in a large population sample. J. Clin. Epidemiol. 59, 465–471 (2006).
Kananen, L. et al. Self-rated health in individuals with and without disease is associated with multiple biomarkers representing multiple biological domains. Sci. Rep. 11, 6139 (2021).
Christian, L. M. et al. Poorer self-rated health is associated with elevated inflammatory markers among older adults. Psychoneuroendocrinology 36, 1495–1504 (2011).
Janszky, I., Lekander, M., Blom, M., Georgiades, A. & Ahnve, S. Self-rated health and vital exhaustion, but not depression, is related to inflammation in women with coronary heart disease. Brain Behav. Immun. 19, 555–563 (2005).
Undén, A.-L. et al. Inflammatory cytokines, behaviour and age as determinants of self-rated health in women. Clin. Sci. (Lond) 112, 363–373 (2007).
Haring, R. et al. Self-perceived quality of life predicts mortality risk better than a multi-biomarker panel, but the combination of both does best. BMC Med. Res. Methodol. 11, 103 (2011).
Barger, S. D., Cribbet, M. R. & Muldoon, M. F. Participant-reported health status predicts cardiovascular and all-cause mortality independent of established and nontraditional biomarkers: Evidence from a representative US sample. J. Am. Heart Assoc. 5, e003741 (2016).
Gupta, S., Xu, Y. & Montgomery, S. The role of inflammation in the relationship of self-rated health with mortality and implications for public health: Data from the English Longitudinal Study of Aging (ELSA). Brain Behav. Immun. Health 8, 100139 (2020).
Rasmussen, L. J. H., Petersen, J. E. V. & Eugen-Olsen, J. Soluble urokinase plasminogen activator receptor (suPAR) as a biomarker of systemic chronic inflammation. Front. Immunol. 12, 5051 (2021).
Eugen-Olsen, J. et al. Circulating soluble urokinase plasminogen activator receptor predicts cancer, cardiovascular disease, diabetes and mortality in the general population. J. Intern. Med. 268, 296–308 (2010).
Haupt, T. H. et al. Healthy lifestyles reduce suPAR and mortality in a Danish general population study. Immun. Age. 16, 77 (2019).
Hayek, S. S. et al. Soluble urokinase receptor and chronic kidney disease. N. Engl. J. Med. 373, 1916–1925 (2015).
Hodges, G. W., Bang, C. N., Wachtell, K., Eugen-Olsen, J. & Jeppesen, J. L. suPAR: A new biomarker for cardiovascular disease?. Can. J. Cardiol. 31, 1293–1302 (2015).
Botha, S. et al. Soluble urokinase plasminogen activator receptor as a prognostic marker of all-cause and cardiovascular mortality in a black population. Int. J. Cardiol. 184, 631–636 (2015).
Rasmussen, L. J. H. et al. Soluble urokinase plasminogen activator receptor (suPAR) in acute care: A strong marker of disease presence and severity, readmission and mortality. A retrospective cohort study. Emerg. Med. J. 33, 769–775 (2016).
Rasmussen, L. J. H. et al. Cumulative childhood risk is associated with a new measure of chronic inflammation in adulthood. J. Child Psychol. Psychiatry 60, 199–208 (2019).
Rasmussen, L. J. H. et al. Association of adverse experiences and exposure to violence in childhood and adolescence with inflammatory burden in young people. JAMA Pediatr. 174, 38–47 (2020).
Rasmussen, L. J. H. et al. Inflammatory biomarkers and cancer: CRP and suPAR as markers of incident cancer in patients with serious nonspecific symptoms and signs of cancer. Int. J. Cancer 141, 191–199 (2017).
Eugen-Olsen, J., Ladelund, S. & Sørensen, L. T. Plasma suPAR is lowered by smoking cessation: A randomized controlled study. Eur. J. Clin. Invest 46, 305–311 (2016).
Hindy, G. et al. Clinical, genetic, and experimental increase in soluble urokinase plasminogen activator receptor levels promotes atherosclerosis. J. Clin. Invest. 2, 555 (2022).
Rasmussen, L. J. H. et al. Association between elevated suPAR, a new biomarker of inflammation, and accelerated aging. J. Gerontol. A Biol. Sci. Med. Sci. 76, 318–327 (2021).
Andreasson, A. N. et al. Inflammation and positive affect are associated with subjective health in women of the general population. J. Health Psychol. 18, 311–320 (2013).
Shanahan, L., Bauldry, S., Freeman, J. & Bondy, C. L. Self-rated health and C-reactive protein in young adults. Brain Behav. Immun. 36, 139–146 (2014).
Leshem-Rubinow, E. et al. Self-rated health is associated with elevated C-reactive protein even among apparently healthy individuals. Isr. Med. Assoc. J. 17, 213–218 (2015).
Sehestedt, T. et al. Soluble urokinase plasminogen activator receptor is associated with subclinical organ damage and cardiovascular events. Atherosclerosis 216, 237–243 (2011).
Lyngbæk, S. et al. Usefulness of soluble urokinase plasminogen activator receptor to predict repeat myocardial infarction and mortality in patients with ST-segment elevation myocardial infarction undergoing primary percutaneous intervention. Am. J. Cardiol. 110, 1756–1763 (2012).
Persson, M. et al. Soluble urokinase plasminogen activator receptor: A risk factor for carotid plaque, stroke, and coronary artery disease. Stroke 45, 18–23 (2014).
Haupt, T. H. et al. Risk factors associated with serum levels of the inflammatory biomarker soluble urokinase plasminogen activator receptor in a general population. Biomark Insights 9, 91–100 (2014).
Quadt, L., Critchley, H. D. & Garfinkel, S. N. The neurobiology of interoception in health and disease. Ann. N. Y. Acad. Sci. 1428, 112–128 (2018).
Dantzer, R. & Kelley, K. W. Twenty years of research on cytokine-induced sickness behavior. Brain Behav. Immun. 21, 153–160 (2007).
Dantzer, R., O’Connor, J. C., Freund, G. G., Johnson, R. W. & Kelley, K. W. From inflammation to sickness and depression: When the immune system subjugates the brain. Nat. Rev. Neurosci. 9, 46–56 (2008).
Weinberger, M. et al. Self-rated health as a predictor of hospital admission and nursing home placement in elderly public housing tenants. Am. J. Public Health 76, 457–459 (1986).
Mulsant, B. H., Ganguli, M. & Seaberg, E. C. The relationship between self-rated health and depressive symptoms in an epidemiological sample of community-dwelling older adults. J. Am. Geriatr. Soc. 45, 954–958 (1997).
Schulz, C.-A. et al. Soluble urokinase-type plasminogen activator receptor (suPAR) and impaired kidney function in the population-based Malmö diet and cancer study. Kidney Int. Rep. 2, 239–247 (2016).
Salanitro, A. H. et al. Inflammatory biomarkers as predictors of hospitalization and death in community-dwelling older adults. Arch. Gerontol. Geriatr. 54, e387-391 (2012).
Cislaghi, B. & Cislaghi, C. Self-rated health as a valid indicator for health-equity analyses: Evidence from the Italian health interview survey. BMC Public Health 19, 533 (2019).
Kyriazopoulou, E. et al. Early treatment of COVID-19 with anakinra guided by soluble urokinase plasminogen receptor plasma levels: A double-blind, randomized controlled phase 3 trial. Nat. Med. 27, 1752–1760 (2021).
Jørgensen, T. et al. A randomized non-pharmacological intervention study for prevention of ischaemic heart disease: Baseline results Inter99. Eur. J. Cardiovasc. Prev. Rehabil. 10, 377–386 (2003).
Jørgensen, T. et al. Effect of screening and lifestyle counselling on incidence of ischaemic heart disease in general population: Inter99 randomised trial. BMJ 348, 777 (2014).
Acknowledgements
The authors would like to thank Prof. Emeritus Torben Jørgensen who designed the Inter99 study, and Thomas Kallemose for statistical assistance.
Author information
Authors and Affiliations
Contributions
H.G.J. and J.E.O. conceived the study in consultation with J.Ta. C.P. collected data. J.Ta. analyzed the data and L.J.H.R., J.P., J.To., J.P., C.P., J.E.O. and H.G.J. made suggestions for analyses. J.Ta. and H.G.J. wrote the first draft of the manuscript, and L.J.H.R., J.To., J.P., J.S., C.P., and J.E.O. contributed to the writing of the manuscript. All authors reviewed and approved the final submission.
Corresponding author
Ethics declarations
Competing interests
J.E.O. is named inventor on patents on suPAR as a prognostic biomarker. The patents are owned by Copenhagen University Hospital Hvidovre, Denmark. J.E.O. is a shareholder, CSO, and co-founder of ViroGates A/S, Denmark. The other authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Tavenier, J., Rasmussen, L.J.H., Tolstrup, J. et al. Self-rated health and chronic inflammation are related and independently associated with hospitalization and long-term mortality in the general population. Sci Rep 12, 19761 (2022). https://doi.org/10.1038/s41598-022-24422-z
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-022-24422-z
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.