Exercise Frequency Reduction Is Associated With Higher Risk of Infection in Newly Diagnosed Diabetes: A Nationally Representative Cohort Study

Yohwan Lim; Hye Jun Kim; Sung Soo Yoon; Sang Jun Lee; Myeong Hoon Lee; Hyewon Park; Sun Jae Park; Seogsong Jeong; Hyun Wook Han

doi:10.3346/jkms.2023.38.e176

Journal List > J Korean Med Sci > v.38(23) > 1516082938

Go to TopGo to Top Go to BottomGo to Bottom

TOOLS

Lim, Kim, Yoon, Lee, Lee, Park, Park, Jeong, and Han: Exercise Frequency Reduction Is Associated With Higher Risk of Infection in Newly Diagnosed Diabetes: A Nationally Representative Cohort Study

Original Article

Preventive & Social Medicine

Journal of Korean Medical Science 2023; 38(23): e176.

Published online: 10 May 2023

DOI: https://doi.org/10.3346/jkms.2023.38.e176

Exercise Frequency Reduction Is Associated With Higher Risk of Infection in Newly Diagnosed Diabetes: A Nationally Representative Cohort Study

Yohwan Lim^1,^2,^*

, Hye Jun Kim^1,^2,^*

, Sung Soo Yoon^1,²

, Sang Jun Lee^1,²

, Myeong Hoon Lee^1,²

, Hyewon Park^1,²

, Sun Jae Park³

, Seogsong Jeong^1,²

, Hyun Wook Han^1,²

¹Department of Biomedical Informatics, CHA University School of Medicine, CHA University, Seongnam, Korea.

²Institute of Biomedical Informatics, CHA University School of Medicine, CHA University, Seongnam, Korea.

³Department of Biomedical Sciences, Seoul National University College of Medicine, Seoul, Korea.

Address for Correspondence: Hyun Wook Han. Department of Biomedical Informatics and Institute of Biomedical Informatics, CHA University School of Medicine, CHA University, 335 Pangyo-ro, Bundang-gu, Seongnam 13488, Republic of Korea. stepano7@gmail.com

Address for Correspondence: Seogsong Jeong. Department of Biomedical Informatics and Institute of Biomedical Informatics, CHA University School of Medicine, CHA University, 335 Pangyo-ro, Bundang-gu, Seongnam 13488, Republic of Korea. seoksong@mensakorea.org

Equal contributor:

^*Yohwan Lim and Hye Jun Kim have contributed equally to this work and share first authorship.

Received 28 August 2022 Accepted 27 February 2023

The Korean Academy of Medical Sciences

https://creativecommons.org/licenses/by-nc/4.0/

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background

Exercise is an important method to control the progression of diabetes. Since diabetes compromises immune function and increases the risk of infectious diseases, we hypothesized that exercise may affect the risk of infection by its immunoprotective effects. However, population-based cohort studies regarding the association between exercise and the risk of infection are limited, especially regarding changes in exercise frequency. The aim of this study was to determine the association between the change in exercise frequency and the risk of infection among patients with newly diagnosed diabetes.

Methods

Data of 10,023 patients with newly diagnosed diabetes were extracted from the Korean National Health Insurance Service-Health Screening Cohort. Self-reported questionnaires for moderate-to-vigorous physical activity (MVPA) were used to classify changes in exercise frequency between two consecutive two-year periods of health screenings (2009–2010 and 2011–2012). The association between changes in exercise frequency and the risk of infection was evaluated using multivariable Cox proportional-hazards regression.

Results

Compared with engaging in ≥ 5 times of MVPA/week during both periods, a radical decrease in MVPA (from ≥ 5 times of MVPA/week to physical inactivity) was associated with a higher risk of pneumonia (adjusted hazard ratio [aHR], 1.60; 95% confidence interval [CI], 1.03–2.48) and upper respiratory tract infection (aHR, 1.15; 95% CI, 1.01–1.31). In addition, a reduction of MVPA from ≥ 5 to < 5 times of MVPA/week was associated with a higher risk of pneumonia (aHR, 1.52; 95% CI, 1.02–2.27), whereas the risk of upper respiratory tract infection was not higher.

Conclusion

Among patients with newly diagnosed diabetes, a reduction in exercise frequency was related to an increase in the risk of pneumonia. For patients with diabetes, a modest level of physical activity may need to be maintained to reduce the risk of pneumonia.

Graphical Abstract

Keywords: Diabetes Mellitus, Exercise, Infection, Cohort Study, Pneumonia

INTRODUCTION

Diabetes mellitus is a chronic illness that affects several organs, including the eyes, gastrointestinal tract, liver, and blood vessels, affecting nearly 6.4% of the global population.1 In South Korea, the prevalence of diabetes was 16.7% in 2020, steadily rising from 11.8% in 2012.2 Lack of exercise, westernized lifestyles, increased body mass index (BMI), small size of pancreatic beta cells, and decreased insulin secretion are thought to contribute to this rising prevalence.3 4

Blood sugar control using oral glucose-lowering medications or insulin injections is the main treatment for diabetes. However, patients should be equally taught about lifestyle modifications, such as engaging in regular exercise, smoking cessation, and weight control, as about medication, as these factors significantly affect disease progression. The general recommendation in terms of physical activity for patients with diabetes is to engage in 150 min or more of moderate-to-vigorous physical activity (MVPA) per week.5 Recently, a cohort study of patients with diabetes in South Korea revealed that those who exercised 5–6 times per week had the lowest mortality rate among those who exercised within the range of 0–7 times per week, emphasizing the positive impact of exercise on the prognosis of this disease.6

However, patients with diabetes reportedly have a higher risk of almost all types of infections, including respiratory infections, than those without diabetes.7 Diabetes compromises a patient’s immunity by reducing T cell-mediated immune responses and impairing neutrophil function. Thus, both the innate and adaptive immune systems are disrupted, thereby increasing the risk of infectious diseases, such as respiratory, urinary tract, soft tissue, and central nervous system infections.8 9

Given that the coronavirus disease 2019 (COVID-19) pandemic has resulted in many individuals being isolated owing to social distancing policies, physical inactivity has become a significant public health concern. However, only a few studies have been conducted to investigate the association between a reduction in exercise and the risk of infectious diseases in patients with diabetes. Therefore, the aim of this study was to examine the association between exercise frequency reduction and risk of infectious diseases among patients with newly diagnosed diabetes.

METHODS

Study population

The Korean National Health Insurance Service (NHIS) provides obligatory health insurance services for all South Korean citizens, covering almost all aspects of healthcare, including biennial health screening, which is carried out for every adult aged at least 40 years. The NHIS collects data on sociodemographic characteristics, medical history, hospitalization and outpatient department visits, serological characteristics, medication prescriptions, and lifestyle behaviors, for research purposes. The study population was derived from the NHIS-Health Screening Cohort, which is a random sample of the entire NHIS database, built with the intent of representing all South Korean adults.

In the present study, patients with newly diagnosed diabetes who underwent consecutive health screenings in two periods (both 2009–2010 and 2011–2012) were identified to evaluate the association between changes in exercise frequency and the risk of infectious disease (n = 33,529). We excluded patients who were diagnosed with infectious diseases before the follow-up investigation (n = 23,263), those for whom information on changes in exercise frequency (n = 183) were missing, and those for whom information of other key variables for the adjusted analyses (n = 60) were missing. Finally, 10,023 patients with newly diagnosed diabetes were included in the analysis (Fig. 1). This study was reported in accordance with the STROBE statement.10

Fig. 1

Flow diagram for the inclusion of study population.

Newly diagnosed diabetes and follow-up for infectious diseases

Patients with newly diagnosed diabetes were identified using the International Classification of Diseases 10th Revision (ICD-10) codes E10-E14 along with the prescription of antidiabetic medications to minimize bias caused by advanced stages of diabetes. Infectious diseases were categorized based on seven different types of organs and identified using previously reported ICD-10 codes (Supplementary Table 1).7 All patients were followed up from the date of the second health screening (carried out from January 1, 2011 to December 31, 2012) until death, diagnosis of infectious disease, or December 31, 2013, whichever occurred earliest.

Changes in exercise frequency

The patients in the NHIS cohort provided responses to self-reported questionnaires on lifestyle behaviors, including moderate physical activity (e.g., dancing, gardening, brisk walking; ≥ 30 min/day) and vigorous physical activity (e.g., running, aerobics, fast cycling; ≥ 20 min/day) per week, at two consecutive, biennial health screenings. Exercise was evaluated using MVPA, which was calculated by adding the frequencies of self-reported moderate physical activity and vigorous physical activity per week. All patients were classified as physically inactive (no MVPA/week), 1–2 times of MVPA/week, 3–4 times of MVPA/week, and ≥ 5 times of MVPA/week. The primary analysis was conducted by examining the change in MVPA between the two periods and its association with the risk of infectious diseases. Each MVPA category for period I (2009–2010) was set as a reference and compared with the MVPA category for period II (2011–2012). Secondary analysis was performed by comparing the patients who continuously engaged in ≥ 5 times/week during both periods with those who had performed ≥ 5 times/week during period I but < 5 times/week during period II, to further examine the association of detraining. The validity and reliability of the physical activity questionnaire have been described previously.11

Key variables

Relevant information was collected for adjusted and stratified analyses, including socioeconomic factors (age [continuous; years], sex [categorical; men or women], and household income [categorical; upper half and lower half]), BMI (continuous; kg/m²), systolic blood pressure (continuous; mmHg), fasting serum glucose (continuous; mg/dL), smoking (categorical; current, former, or never), alcohol consumption (categorical; yes or no), and comorbidities (continuous; Charlson Comorbidity Index [CCI]). Household income was used as a proxy for insurance premiums. The CCI was calculated as described in a previous study.12

Statistical analysis

Descriptive characteristics are presented as number (%) and median (interquartile range; IQR) for categorical and continuous variables, respectively. The crude rate was calculated based on the total number of events per 1,000 person-years (PY) for each group. The hazard ratio (HR) and 95% confidence interval (CI) were calculated using Cox proportional-hazards regression. The first model was minimally adjusted for age and sex. The adjusted HR (aHR) and 95% CI of the final model were calculated after adjusting for key variables including age, sex, household income, BMI, systolic blood pressure, fasting serum glucose, smoking, alcohol consumption, and CCI. The proportional-hazards assumption was graphically tested using a log-log plot. Restricted cubic splines were generated based on three knots to present the consecutive aHR (95% CI) according to changes in MVPA frequency. Two-sided P values less than 0.05 were considered statistically significant. All statistical analyses were performed using SAS (version 9.4; SAS Institute, Cary, NC, USA) and R (version 3.4.0, 2017; R Development Core Team; Vienna, Austria).

Ethics statement

This study was approved by the Institutional Review Board of CHA University Hospital (No.: CHAMC 2022-04-041). The requisite for informed consent was waived because the data was provided in an anonymized form according to the strict confidentiality guidelines.

RESULTS

Study population characteristics

Among the 10,023 patients with newly diagnosed diabetes, the median age was 62 years (IQR, 55–70) in period II. The proportions of all categories of exercise frequency were similar in both periods. In period I, 48.0% (n = 4,815) of participants were physically inactive, followed by ≥ 5 (n = 2,438; 24.3%), 1–2 (n = 1,469; 14.7%), and 3–4 (n = 1,301; 13.0%) times of MVPA/week. Similarly, approximately half of the patients (n = 4,903; 48.9%) were physically inactive during period II, followed by ≥ 5 (n = 2,446; 24.4%), 1–2 (n = 1,418; 14.1%), and 3–4 (n = 1,256; 12.5%) times of MVPA/week. All the descriptive characteristics of the patients are summarized in Table 1.

Table 1

Baseline characteristics of the study population

Characteristics			Total (N = 10,023)
Health screening period I (2009–2010)
	Exercise frequency
		Physically inactive	4,815 (48.0)
		1–2 times of MVPA/week	1,469 (14.7)
		3–4 times of MVPA/week	1,301 (13.0)
		≥ 5 times of MVPA/week	2,438 (24.3)
Health screening period II (2011–2012)
	Exercise frequency
		Physically inactive	4,903 (48.9)
		1–2 times of MVPA/week	1,418 (14.1)
		3–4 times of MVPA/week	1,256 (12.5)
		≥ 5 times of MVPA/week	2,446 (24.4)
Age, yr			62 (55–70)
Household income^a
	Lower half		3,688 (36.8)
	Upper half		6,335 (63.2)
Body mass index, kg/m²			24.5 (22.6–26.4)
Waist circumference, cm			85 (80–90)
Systolic blood pressure, mmHg			130 (120–138)
Diastolic blood pressure, mmHg			80 (70–83)
Fasting serum glucose, mg/dL			127 (108–152)
Cigarette smoking
	Never smoker		5,560 (55.5)
	Former smoker		2,304 (23.0)
	Current smoker		2,159 (21.5)
Alcohol consumption
	Yes		4,120 (41.1)
	No		5,903 (58.9)
Charlson comorbidity index
	0		1,492 (14.9)
	1		1,778 (17.7)
	≥ 2		6,753 (67.4)

Continuous variables were presented as median (interquartile range) and categorical variables as number (%) unless otherwise specified.

MVPA = moderate-to-vigorous physical activity.

^aProxy for socioeconomic status based on the insurance premium of the National Health Insurance Service.

Changes in MVPA frequency and risk of infection

Compared to continuous engagement in ≥ 5 times of MVPA/week during both periods, a radical decrease in MVPA frequency (from ≥ 5 times of MVPA/week to a state of physical inactivity) was associated with a higher risk of pneumonia (aHR, 1.60; 95% CI, 1.03–2.48) and upper respiratory tract infection (aHR, 1.15; 95% CI, 1.01–1.31; Table 2). In addition, a reduction in MVPA from ≥ 5 to 3–4 times/week was associated with a higher risk of upper urinary tract infection (aHR, 3.49; 95% CI, 1.22–10.02); however, the wide CI indicates an imprecise estimate owing to the relatively small infection events in these groups (n = 28).

Table 2

Association of exercise frequency reduction with risk of infection

Type of infection		Exercise frequency during period II (2011–2012)				P for trend
Type of infection		≥ 5 times of MVPA/week	3–4 times of MVPA/week	1–2 times of MVPA/week	Physically inactive	P for trend
≥ 5 times of MVPA/week during period I (2009–2010)
	Gastrointestinal infection	1.00 (reference)	0.88 (0.62–1.25)	0.77 (0.52–1.15)	0.91 (0.70–1.18)	0.581
	Upper respiratory tract infection	1.00 (reference)	0.93 (0.78–1.12)	0.90 (0.73–1.09)	1.15 (1.01–1.31)^a	0.032
	Pneumonia	1.00 (reference)	1.31 (0.72–2.40)	1.57 (0.84–2.92)	1.60 (1.03–2.48)^a	0.194
	Biliary tract infection	1.00 (reference)	2.46 (0.98–6.16)	0.80 (0.78–3.64)	1.80 (0.80–4.05)	0.175
	Skin and soft tissue infection	1.00 (reference)	1.15 (0.88–1.50)	0.91 (0.67–1.25)	1.15 (0.94–1.41)	0.339
	Upper urinary tract infection	1.00 (reference)	3.49 (1.22–10.02)^a	1.82 (0.47–7.09)	1.45 (0.55–3.85)	0.123
	Lower urinary tract infection	1.00 (reference)	1.12 (0.74–1.69)	1.13 (0.73–1.76)	0.98 (0.72–1.34)	0.883
3–4 times of MVPA/week during period I (2009–2010)
	Gastrointestinal infection	1.44 (0.90–2.29)	1.00 (reference)	0.83 (0.51–1.37)	0.87 (0.57–1.32)	0.434
	Upper respiratory tract infection	1.01 (0.80–1.28)	1.00 (reference)	1.12 (0.86–1.46)	1.19 (0.94–1.50)	0.547
	Pneumonia	1.51 (0.66–3.44)	1.00 (reference)	0.76 (0.30–1.89)	1.01 (0.48–2.12)	0.903
	Biliary tract infection	2.56 (0.26–24.90)	1.00 (reference)	2.28 (0.41–12.60)	1.16 (0.22–6.30)	0.435
	Skin and soft tissue infection	1.11 (0.77–1.62)	1.00 (reference)	1.17 (0.78–1.74)	1.04 (0.73–1.48)	0.505
	Upper urinary tract infection	2.49 (0.78–12.99)	1.00 (reference)	NA	1.77 (0.19–16.69)	0.555
	Lower urinary tract infection	0.78 (0.36–1.69)	1.00 (reference)	1.05 (0.55–1.99)	0.75 (0.42–1.34)	0.655
1–2 times of MVPA/week during period I (2009–2010)
	Gastrointestinal infection	1.44 (0.90–2.29)	1.16 (0.70–1.93)	1.00 (reference)	1.29 (0.85–1.95)	0.461
	Upper respiratory tract infection	1.01 (0.80–1.28)	0.98 (0.77–1.26)	1.00 (reference)	0.95 (0.77–1.16)	0.916
	Pneumonia	1.51 (0.66–3.44)	0.85 (0.32–2.31)	1.00 (reference)	1.15 (0.54–2.44)	0.582
	Biliary tract infection	2.56 (0.26–24.90)	3.45 (0.35–33.80)	1.00 (reference)	4.28 (0.55–33.31)	0.511
	Skin and soft tissue infection	1.11 (0.77–1.62)	1.24 (0.85–1.82)	1.00 (reference)	1.13 (0.82–1.56)	0.737
	Upper urinary tract infection	2.49 (0.78–12.99)	NA	1.00 (reference)	2.91 (0.64–13.21)	0.592
	Lower urinary tract infection	0.78 (0.36–1.69)	1.29 (0.62–2.68)	1.00 (reference)	1.46 (0.81–2.62)	0.190
Physically inactive during period I (2009–2010)
	Gastrointestinal infection	1.14 (0.93–1.40)	1.07 (0.82–1.40)	0.85 (0.66–1.09)	1.00 (reference)	0.245
	Upper respiratory tract infection	1.04 (0.93–1.16)	1.07 (0.93–1.23)	1.03 (0.92–1.16)	1.00 (reference)	0.764
	Pneumonia	0.99 (0.70–1.39)	1.29 (0.86–1.94)	1.02 (0.71–1.48)	1.00 (reference)	0.658
	Biliary tract infection	0.82 (0.40–1.68)	0.94 (0.37–2.38)	0.77 (0.35–1.72)	1.00 (reference)	0.896
	Skin and soft tissue infection	1.02 (0.86–1.22)	1.17 (0.95–1.45)	1.22 (1.02–1.45)^a	1.00 (reference)	0.100
	Upper urinary tract infection	0.91 (0.45–1.87)	1.05 (0.45–2.46)	1.30 (0.67–2.52)	1.00 (reference)	0.857
	Lower urinary tract infection	0.94 (0.72–1.23)	1.27 (0.94–1.71)	0.81 (0.60–1.09)	1.00 (reference)	0.162

Data are adjusted hazard ratio calculated using the Cox proportional hazards model after adjustments for age, sex, household income, body mass index, systolic blood pressure, fasting serum glucose, smoking, alcohol consumption, and Charlson Comorbidity Index.

MVPA = moderate-to-vigorous physical activity, NA = not applicable.

^aP < 0.05.

Compared to maintained physical inactivity during both periods, an increase in MVPA from a state of physical inactivity to 1–2 times/week was associated with a higher risk of skin and soft tissue infection (aHR, 1.22; 95% CI, 1.02–1.45). However, no infectious diseases significantly increased when MVPA radically increased from a state of physical inactivity to ≥ 5 times/week.

Table 3 summarizes the results of the secondary analysis of the risk of pneumonia and upper respiratory tract infection, which showed a statistical significance at the primary analysis when MVPA radically decreased. The crude rates of these diseases per 1,000 PY of those who engaged in ≥ 5 times of MVPA/week during period II (2011–2012) and those who performed < 5 times of MVPA/week in that period among participants engaged in ≥ 5 times of MVPA/week during period I (2009–2010) were 16.2 and 27.6, respectively, for pneumonia, and 344.2 and 374.1, respectively, for upper respiratory tract infection. A reduction in MVPA frequency was associated with a higher risk of pneumonia (aHR, 1.52; 95% CI, 1.02–2.27), but not with that of upper respiratory tract infection. Other infectious diseases were analyzed in the same manner but did not yield significant results when the MVPA frequency was reduced (Supplementary Table 2).

Table 3

Association of exercise frequency reduction with risk of infection among participants with ≥ 5 times of MVPA/week during period I (2009–2010)

Type of infection		≥ 5 times of MVPA/week during period I (2009–2010)		P value
Type of infection		≥ 5 times of MVPA/week during period II (2011–2012) (n = 1,083)	< 5 times of MVPA/week during period II (2011–2012) (n = 1,355)	P value
Pneumonia
	Event (%)	37 (3.4)	81 (6.0)
	PY	2,288.50	2,931.4
	Crude rate/1,000 PY	16.2	27.6
	Unadjusted HR (95% CI)	1.00 (reference)	1.71 (1.16–2.53)	0.007
	aHR^a (95% CI)	1.00 (reference)	1.59 (1.07–2.36)	0.021
	aHR^b (95% CI)	1.00 (reference)	1.52 (1.02–2.27)	0.040
Upper respiratory tract infection
	Event (%)	519 (47.9)	717 (52.9)
	PY	1,507.8	1,916.80
	Crude rate/1,000 PY	344.2	374.1
	Unadjusted HR (95% CI)	1.00 (reference)	1.09 (0.98–1.22)	0.127
	aHR^a (95% CI)	1.00 (reference)	1.03 (0.92–1.16)	0.572
	aHR^b (95% CI)	1.00 (reference)	1.04 (0.93–1.17)	0.515

Data are adjusted hazard ratio calculated using the Cox proportional hazards.

MVPA = moderate-to-vigorous physical activity, PY = person-years, HR = hazard ratio, aHR = adjusted hazard ratio, CI = confidence interval.

^aaHR adjusted for age and sex.

^baHR adjusted for age, sex, household income, body mass index, systolic blood pressure, fasting serum glucose, smoking, alcohol consumption, and Charlson Comorbidity Index.

Restricted cubic splines were generated for the presentation of consecutive aHR and 95% CI with the risk of pneumonia and upper respiratory tract infection according to the changes in MVPA. As illustrated in Fig. 2, the risk of pneumonia was significantly elevated when the frequency of MVPA decreased from ≥ 5 times of MVPA/week in period I to < 5 times of MVPA/week in period II. However, no significant associations were discovered with the risk of upper respiratory tract infection (Supplementary Fig. 1).

Fig. 2

Restricted cubic spline evaluating the association of MVPA frequency during period II (2011–2012) with the risk of pneumonia among patients with ≥ 5 times of MVPA/week during period I (2009–2010). aHR was calculated using Cox proportional hazards regression after adjustments for age, sex, household income, body mass index, systolic blood pressure, fasting serum glucose, smoking, alcohol consumption, and Charlson Comorbidity Index. Blue solid line indicates aHR and beige shaded area indicates 95% CIs.

MVPA = moderate-to-vigorous physical activity, aHR = adjusted hazard ratio, CI = confidence interval.

DISCUSSION

In this study, the risks of both pneumonia and upper respiratory tract infection increased when MVPA frequency was reduced from ≥ 5 times of MVPA/week to a state of physical inactivity. However, a reduction in MVPA frequency from ≥ 5 to < 5 times/week only increased the risk of pneumonia, emphasizing the role of maintaining regular exercise in pneumonia. Therefore, being physically active may provide additional benefits to patients with diabetes regarding their future risks of pneumonia.

Our results showed a limited effect of MVPA in that they demonstrated only the detrimental effects of a reduction in MVPA: an increased risk of pneumonia. Many previous studies have demonstrated the beneficial effects of regular exercise on the risk of pneumonia. A systematic review and meta-analysis of 10 prospective studies involving 1,044,492 participants revealed strong evidence for a lower risk of pneumonia among the most physically active group than that in the least physically active group (relative risk, 0.69; 95% CI, 0.64–0.74).13 Moreover, a randomized controlled trial of 154 US adults aged ≥ 50 years revealed that 8 weeks of MVPA was effective in preventing acute respiratory illness.14 However, few studies have been conducted to examine the effects of exercise reduction and physical inactivity on the risk of pneumonia. A cohort study of 48,440 adult patients diagnosed with COVID-19 revealed that patients with COVID-19 who were consistently physically inactive had a greater risk of hospitalization (odds ratio [OR], 2.26; 95% CI, 1.81–2.83), intensive care unit admission (OR, 1.73; 95% CI, 1.18–2.55), and mortality rate (OR, 2.49; 95% CI, 1.33–4.67) than those who were consistently physically active, which was similar to the results in our study: detraining can have detrimental effects in the pneumonia risk.15 MVPA levels were separated into three categories (consistently inactive, inconsistently active, and consistently active) in the study by Sallis et al.,15 based on self-reported exercise vital signs, which was similar to what we did in this study. In addition, key variables that were used for adjustment included chronic diseases, which could represent current conditions on exercise, which were similar to our study. However, Sallis et al.15 targeted the severe outcomes of COVID-19 which was different from the case in our study, considering that we mainly targeted the incidence of infectious diseases. Another cohort study of 1,365 women revealed a significantly high risk of infection that precipitates hospitalization when physically inactive (OR, 4.08; 95% CI, 1.73–9.63).16 Among the infectious diseases in the study by Leveille et al.,16 pneumonia was the most frequent (33%) during the follow-up, which is similar to our results. However, the authors analyzed the risk of infections in general and did not conduct further analysis to evaluate the risk of pneumonia itself, which made it difficult to compare with our results.

Patients with diabetes have a higher risk of immune dysfunction owing to depressed polymorphonuclear leukocyte function.17 Moreover, several studies have revealed that detraining leads to impaired insulin sensitivity, as tested using the euglycemic clamp technique.18 Therefore, reducing exercise may result in the progression of diabetes in a poorly controlled state, leading to immune dysfunction and an increased risk of pneumonia, as it is the most common cause of hospitalization among patients with diabetes.7 19 However, we could not detect any dose-response relationship between MVPA and risk of respiratory infections, which is inconsistent with the results of previous studies.20 21 22 The main difference between this study and prior studies is the measurement of physical activity. In most studies in which such a dose-response relationship was demonstrated, physical activity was measured as the metabolic equivalent of tasks (METs),21 22 whereas physical activity measured using self-reported questionnaires had a direct but not dose-response relationship between physical inactivity and increased risk of respiratory infections.16 When measuring physical activity obtained with self-reported questionnaires, assessment of moderate-intensity physical activity is less robust than assessment of heavy-intensity physical activity owing to its susceptibility to recall bias.23 24 Therefore, the inconsistency between the reduction in exercise and the risk of respiratory infections requires further investigation.

Our results did not include a significant relationship between the risk of upper respiratory tract infection and a reduction in MVPA, except for a reduction due to physical inactivity. As our data were medical claims data, upper respiratory tract infections may have been underestimated owing to relatively mild symptoms and the use of over-the-counter pharmaceuticals. As most previous studies regarding the risk of upper respiratory tract infections also made use of self-reported data,14 20 22 the true association might have been confounded because of response bias.

The risk of skin and soft tissue infections increased when the patients became more active compared with those who were physically inactive. Physical inactivity caused by a decreased general condition or other disease states might have increased skin infection owing to poor personal hygiene, even though we adjusted the risk with the CCI score. In addition, exercise itself may be a risk factor for skin infection owing to the sweaty environment and high skin contact.25

The reason why an increase in MVPA did not result in a decrease in the risk of infection and was inconsistent with previous studies in which exercise-induced, immunoprotective effects on the risk of respiratory infections were demonstrated, is unclear.13 14 20 26 27 The main confounder responsible for this inconsistency is possibly the MVPA level prior to the infection, as suggested in prior studies.21 We attempted to reflect the trend of each individual’s MVPA level by comparing the previous MVPA frequency with the latest (changes in MVPA from periods I to II), self-reported during biennial health screening tests, which holds promise in minimizing reverse causality of MVPA frequency. However, owing to the nature of this retrospective, observational cohort study, the MVPA frequency just before the infection is unknown, and the reported MVPA frequency could have changed after the health screening test. As short-term physical inactivity can lead to deterioration of insulin resistance and immune functions,28 29 30 the MVPA frequency before infection is likely to bias the true association. As we focused on investigating the long-term effect of changes in MVPA frequency on the risk of infection, further studies are required to investigate the combined short- and long-term effects of MVPA frequency on the risk of infection. Other possible reasons accounting for the insignificant results are heavy exertion or repeated exercise that may cause transient immune dysfunction, inadequate statistical power of specific infections, or unmeasured confounders that account for the lack of relationships between changes in MVPA frequencies and infection risk.26 31

To the best of our knowledge, this is the first cohort study in which the infection risk was compared among patients, newly diagnosed with diabetes, with different exercise frequencies. In addition, we discovered evidence of detrimental effects of reduced exercise frequency on the risk of infection. Owing to the COVID-19 pandemic, social distancing and mass quarantine have led to increased concerns about physical inactivity and its consequences on global public health.32 Our results highlight the clinical impact and health consequences of physical inactivity and detraining among patients with diabetes. However, this study had several limitations. First, as only a few studies have been conducted to analyze the effect of detraining on infection risk, we were unable to compare our results with those of previous studies. Second, the self-reported questionnaires provided insufficient information on MVPA duration, intensity, and type (e.g., aerobic, resistance, and combined). Future studies should make use of METs, which are more accurate in the evaluation of MVPA. Third, owing to the retrospective nature of the study, the cause-effect relationship could not be established with certainty. Fourth, we could not divide the study population into those with type 1 and type 2 diabetes because of the retrospective operational definitions for diabetes. The proposed mechanisms of detrimental effects of detraining are based on changes in insulin sensitivity, which are incompatible with type 1 diabetes. As the main pathophysiology of type 1 diabetes is immune-mediated destruction of pancreatic beta cells, different results might have been obtained if such subgroup analysis were performed.33 However, given the low prevalence of type 1 diabetes among South Korean adults (approximately 0.02%), our results may be applicable to patients with type 2 diabetes.34 Finally, potential confounders might have remained in our analyses owing to the lack of detailed clinical information (e.g., dietary information and total energy intake). Even though our study included patients with newly diagnosed diabetes and was adjusted for key variables, diabetes-related complications were not sufficiently evaluated owing to the lack of detailed clinical information for diabetes, such as results of the Michigan Neuropathy Screening Instrument and carotid Doppler ultrasonography.

In conclusion, a reduction in MVPA frequency was associated with an increased risk of pneumonia among patients with newly diagnosed diabetes. For patients with diabetes, a modest level of physical activity may need to be maintained to reduce the risk of pneumonia. Further research is needed to determine the incremental and decremental effects of type-specific exercise on infection risk among patients with diabetes.

Notes

Funding: This research was supported by a grant of the MD-Phd/Medical Scientist Training Program through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea. This work was supported by the Bio Industry Technology Development Program (No. 20015086) of the Ministry of Trade, Industry & Energy (MOTIE, Korea), and a grant of the Information and Communications Promotion Fund through the National IT Industry Promotion Agency (NIPA), funded by the Ministry of Science and ICT (MSIT), Republic of Korea.

Disclosure: The authors have no potential conflicts of interest to disclose.

Data Availability Statement: No additional data available. Authors were granted to access the National Health Insurance Service (NHIS) database by registering at Big Data Steering Department in the Republic of Korea.

Author Contributions:

Conceptualization: Lim Y, Kim HJ, Yoon SS, Lee SJ, Jeong S, Han HW.
Data curation: Lim Y, Kim HJ, Lee MH, Park H, Park SJ, Jeong S, Han HW.
Formal analysis: Lim Y, Kim HJ, Jeong S.
Methodology: Lim Y, Kim HJ, Yoon SS, Lee SJ, Lee MH, Park H, Park SJ, Jeong S, Han HW.
Supervision: Han HW.
Writing - original draft: Lim Y, Kim HJ, Jeong S, Han HW.
Writing - review & editing: Lim Y, Kim HJ, Yoon SS, Lee SJ, Lee MH, Park H, Park SJ, Jeong S, Han HW.

References

1. Shaw JE, Sicree RA, Zimmet PZ. Global estimates of the prevalence of diabetes for 2010 and 2030. Diabetes Res Clin Pract. 2010; 87(1):4–14. PMID: 19896746.

2. Bae JH, Han KD, Ko SH, Yang YS, Choi JH, Choi KM, et al. Diabetes fact sheet in Korea 2021. Diabetes Metab J. 2022; 46(3):417–426. PMID: 35656565.

3. Lee Y, Lim H, Park S. Characteristics of non-obese diabetes in Korean population. Public Health Wkly Rep. 2017; 10(30):780–787.

4. Kwon H, Kim W, Lee D, Park S. Current status and need for a Korean diabetes prevention study. Public Health Wkly Rep. 2015; 8:746–753.

5. American Diabetes Association. 5. Lifestyle management: Standards of Medical Care in Diabetes-2019. Diabetes Care. 2019; 42(Suppl 1):S46–S60. PMID: 30559231.

6. Shin WY, Lee T, Jeon DH, Kim HC. Diabetes, frequency of exercise, and mortality over 12 years: analysis of the National Health Insurance Service-Health Screening (NHIS-HEALS) database. J Korean Med Sci. 2018; 33(8):e60. PMID: 29441753.

7. Kim EJ, Ha KH, Kim DJ, Choi YH. Diabetes and the risk of infection: a national cohort study. Diabetes Metab J. 2019; 43(6):804–814. PMID: 31701687.

8. Casqueiro J, Casqueiro J, Alves C. Infections in patients with diabetes mellitus: a review of pathogenesis. Indian J Endocrinol Metab. 2012; 16:Suppl 1. (Suppl1):S27–S36. PMID: 22701840.

9. Seo JA. The Effects of Diabetes on Risk of Infection. J Korean Diabetes. 2017; 18(1):14–19.

10. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ. 2007; 335(7624):806–808. PMID: 17947786.

11. Jeong HG, Kim DY, Kang DW, Kim BJ, Kim CK, Kim Y, et al. Physical activity frequency and the risk of stroke: a nationwide cohort study in Korea. J Am Heart Assoc. 2017; 6(9):e005671. PMID: 28855168.

12. Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005; 43(11):1130–1139. PMID: 16224307.

13. Kunutsor SK, Seidu S, Laukkanen JA. Physical activity reduces the risk of pneumonia: systematic review and meta-analysis of 10 prospective studies involving 1,044,492 participants. Geroscience. 2022; 44(1):519–532. PMID: 34822066.

14. Barrett B, Hayney MS, Muller D, Rakel D, Ward A, Obasi CN, et al. Meditation or exercise for preventing acute respiratory infection: a randomized controlled trial. Ann Fam Med. 2012; 10(4):337–346. PMID: 22778122.

15. Sallis R, Young DR, Tartof SY, Sallis JF, Sall J, Li Q, et al. Physical inactivity is associated with a higher risk for severe COVID-19 outcomes: a study in 48 440 adult patients. Br J Sports Med. 2021; 55(19):1099–1105. PMID: 33849909.

16. Leveille SG, Gray S, LaCroix AZ, Ferrucci L, Black DJ, Guralnik JM. Physical inactivity and smoking increase risk for serious infections in older women. J Am Geriatr Soc. 2000; 48(12):1582–1588. PMID: 11129746.

17. Hirata Y, Tomioka H, Sekiya R, Yamashita S, Kaneda T, Kida Y, et al. Association of hyperglycemia on admission and during hospitalization with mortality in diabetic patients admitted for pneumonia. Intern Med. 2013; 52(21):2431–2438. PMID: 24190147.

18. Burstein R, Polychronakos C, Toews CJ, MacDougall JD, Guyda HJ, Posner BI. Acute reversal of the enhanced insulin action in trained athletes. Association with insulin receptor changes. Diabetes. 1985; 34(8):756–760. PMID: 3894119.

19. Martins M, Boavida JM, Raposo JF, Froes F, Nunes B, Ribeiro RT, et al. Diabetes hinders community-acquired pneumonia outcomes in hospitalized patients. BMJ Open Diabetes Res Care. 2016; 4(1):e000181.

20. Nieman DC, Sakaguchi CA. Physical activity lowers the risk for acute respiratory infections: Time for recognition. J Sport Health Sci. 2022; 11(6):648–655. PMID: 35995362.

21. Williams PT. Dose-response relationship between exercise and respiratory disease mortality. Med Sci Sports Exerc. 2014; 46(4):711–717. PMID: 24002349.

22. Fondell E, Lagerros YT, Sundberg CJ, Lekander M, Bälter O, Rothman KJ, et al. Physical activity, stress, and self-reported upper respiratory tract infection. Med Sci Sports Exerc. 2011; 43(2):272–279. PMID: 20581713.

23. Jacobs DR Jr, Ainsworth BE, Hartman TJ, Leon AS. A simultaneous evaluation of 10 commonly used physical activity questionnaires. Med Sci Sports Exerc. 1993; 25(1):81–91. PMID: 8423759.

24. Sylvia LG, Bernstein EE, Hubbard JL, Keating L, Anderson EJ. Practical guide to measuring physical activity. J Acad Nutr Diet. 2014; 114(2):199–208. PMID: 24290836.

25. Pujalte GG, Costa LM, Clapp AD, Presutti RJ, Sluzevich JC. More than skin deep: dermatologic conditions in athletes. Sports Health. 2023; 15(1):74–85. PMID: 35034516.

26. Nieman DC, Wentz LM. The compelling link between physical activity and the body’s defense system. J Sport Health Sci. 2019; 8(3):201–217. PMID: 31193280.

27. Nieman DC. Exercise is medicine for immune function: implication for COVID-19. Curr Sports Med Rep. 2021; 20(8):395–401. PMID: 34357885.

28. Alibegovic AC, Højbjerre L, Sonne MP, van Hall G, Stallknecht B, Dela F, et al. Impact of 9 days of bed rest on hepatic and peripheral insulin action, insulin secretion, and whole-body lipolysis in healthy young male offspring of patients with type 2 diabetes. Diabetes. 2009; 58(12):2749–2756. PMID: 19720789.

29. Hoff P, Belavý DL, Huscher D, Lang A, Hahne M, Kuhlmey AK, et al. Effects of 60-day bed rest with and without exercise on cellular and humoral immunological parameters. Cell Mol Immunol. 2015; 12(4):483–492. PMID: 25382740.

30. Booth FW, Roberts CK, Laye MJ. Lack of exercise is a major cause of chronic diseases. Compr Physiol. 2012; 2(2):1143–1211. PMID: 23798298.

31. Murphy EA, Davis JM, Carmichael MD, Gangemi JD, Ghaffar A, Mayer EP. Exercise stress increases susceptibility to influenza infection. Brain Behav Immun. 2008; 22(8):1152–1155. PMID: 18616997.

32. Kim EA. Social distancing and public health guidelines at workplaces in Korea: responses to coronavirus disease-19. Saf Health Work. 2020; 11(3):275–283. PMID: 32837740.

33. Atkinson MA, Eisenbarth GS, Michels AW. Type 1 diabetes. Lancet. 2014; 383(9911):69–82. PMID: 23890997.

34. Song SO, Song YD, Nam JY, Park KH, Yoon JH, Son KM, et al. Epidemiology of type 1 diabetes mellitus in Korea through an investigation of the national registration project of type 1 diabetes for the reimbursement of glucometer strips with additional analyses using claims data. Diabetes Metab J. 2016; 40(1):35–45. PMID: 26912154.

SUPPLEMENTARY MATERIALS

Supplementary Table 1

ICD-10 diagnostic codes for infectious diseases

jkms-38-e176-s001.doc

Supplementary Table 2

Association of exercise frequency reduction with risk of infection among participants with ≥ 5 times of MVPA/week during period I (2009–2010)

jkms-38-e176-s002.doc

Supplementary Fig. 1

Restricted cubic spline evaluating the association of MVPA frequency during period II (2011–2012) with the risk of upper respiratory tract infection among patients with ≥ 5 times of MVPA/week during period I (2009–2010). aHR was calculated using Cox proportional hazards regression after adjustments for age, sex, household income, body mass index, systolic blood pressure, fasting serum glucose, smoking, alcohol consumption, and Charlson Comorbidity Index. Blue solid line indicates aHR and beige shaded area indicates 95% CIs.

jkms-38-e176-s003.doc

TOOLS

Similar articles