Association between Korean Medicine Hospital Utilization and Cardiovascular Risks in Patients with Hypertension: a National Korean Cohort Study

Article information

J Korean Med. 2019;40(3):1-20
Publication date (electronic) : 2019 September 30
doi : https://doi.org/10.13048/jkm.19024
1Department of Health Science, Graduate School of Applied Korean Medicine, Kyung Hee University
2Department of Meridian and Acupoint, Graduate School of Science in Korean Medicine, Kyung Hee University
Correspondence to: (Sabina Lim), Graduate School of Applied Korean Medicine, and Graduate School of Science in Korean Medicine, Kyung Hee University, 26 Kyungheedae-ro, Dongdaemun-gu, Seoul 02447, Korea, Tel: +82-2-961-0324, Fax: 0504-072-7831, E-mail: lims@khu.ac.kr
Received 2019 June 3; Revised 2019 July 25; Accepted 2019 July 26.

Abstract

Objectives

This study aims to investigate the effects of Korean Medicine Hospital Utilization (KMHU) on major adverse cardiovascular events (MACE), myocardial infarction (MI), stroke, and death in hypertensive patients taking antihypertensives.

Methods

Using the Korean National Health Insurance Service-National Sample Cohort database, this study identified and diagnosed 68,457 hypertensive patients taking antihypertensives between 2003 and 2006. They were divided into KMHU and non-KMHU groups. The follow-up period ended with the diagnosis of myocardial infarction, stroke, or death. After propensity score matching (PSM), there were 18,242 patients each in the non-KMHU and KMHU groups. We calculated the incidence rate, hazard ratio (HR), and 95% confidence interval (CI) for MACE, myocardial infarction, stroke, and death in patients with hypertension using a stratified Cox proportional hazard model. In addition, secondary outcome analyses for stroke and cardiovascular mortality were performed.

Results

After PSM, the HRs for MACE (HR: 0.84, 95% CI: 0.81–0.87), all-cause mortality (HR: 0.75, 95% CI: 0.72–0.79), and myocardial infarction (HR: 0.90, 95% CI: 0.83–0.97) were significantly lower in the KMHU group than in the non-KMHU group. Moreover, the HRs for stroke-related mortality, haemorrhage and ischaemic stroke-related mortality, and ischaemic heart disease-related and circulatory system disease-related mortality were significantly lower in the KMHU group than in the non-KMHU group.

Conclusions

On long-term follow-up observation, this study supported the effect of KMHU for managing hypertension and reducing the burden of cardiovascular diseases.

Fig. 1

Flowchart of sample selection process. From 1 million sampled data, patients who were newly diagnosed with hypertension, between 2003 and 2006, were included in this study. Except for subjects that did not meet the criteria, the korean medicine hospital utilization group and the non-korean medicine hospital utilization group group matched with propensity score at a 1:1 ratio.

Fig. 2

Kaplan-Meier Curves on MACE, all-cause mortality, myocardial infarction, and stroke in Korean medicine hospital utilization group and non-Korean medicine hospital utilization group before PSM, after PSM and after stabilized IPTW. MACE indicates Major Adverse Cardiovascular Events; PSM, Propensity Score Matching; and s IPTW, Stabilized Inverse Probability of Treatment Weighting.

Fig. 3

Kaplan-Meier curves on cerebro-cardio vascular mortality in Korean medicine hospital utilization group and non-Korean medicine hospital utilization group before PSM, after PSM and after stabilized IPTW. CSD indicates Circulatory System Disease ; IHD, Ischemic Heart Disease; PSM, Propensity Score Matching; and s IPTW, stabilized Inverse Probability of Treatment Weighting.

Incidence rate of MACE, all-cause mortality, myocardial infarction, and stroke by acupuncture

Hazard ratios of MACE, all-cause mortality, myocardial infarction and stroke treated with acupuncture

Incidence rate of cerebro-cardiovascular mortality by acupuncture

Hazard ratios of cerebro-cardiovascular mortality by acupuncture

References

1. Kearney PM, Whelton M, Reynolds K, Muntner P, Whelton PK, He J. Global burden of hypertension: analysis of worldwide data. The Lancet 2005;365(9455):217–23.
2. Korea Centers for Disease Control and Prevention. The Seventh Korea National Health and Nutrition Examination Survey (VII-1 2016) 2016.
3. The Korean Society of Hypertension. Guidelines for Hypertension Treatment in 2018 2018;:10–1.
4. Ministry of Health & Welfare. Status and development direction of chronic disease management business Lee JG (Ministry of Health & Welfare Health Policy Officer).
5. Ministry of Health and Welfare in Government of the Republic of Korea. National Health Plan 2010. Accessed April 25, 2019. https://www.khealth.or.kr/board?menuId=MENU00725&siteId=null .
6. Ministry of Health & Welfare Korea Institute for Health and Social Affairs. Comprehensive national health promotion plan(HP2010) Interim Evaluation and Amendment Report 2006.
7. Korea Institute for Health and Social Affairs. Medical Unification Plan’ for shared growth with modern medicine and traditional medicine in Korea. Occasional report 2013;:21–2.
8. Jeong WY. The brief history of Korean traditional medicine (1899–1999). Korean J Med Hist 1999;8(2):169–255.
9. Republic of Korea. Economic bulletin January 2000 Issue 74–77 Korea Development Institute.
10. Korean Medicine Hospitals’ Association. Korean Classification of Diseases (한의) Practical application training HAN CH; 2009.
11. Ministry of Government Legislation. Medical assistant law commentary 1963. p. 08.
12. Korea Health Industry Development Institute. Korea Institute of Science & Technology Evaluation and Planning R&D FOCUS. Korean medicine R&D Long-Term Development Plan Study 2006. p. 55–63.
13. Korea Institute of Science & Technology Evaluation and Planning. Advancement of oriental medicine R & D
14. Lee BH, Kim CH, Seo JC, Youn HM, Jang KJ, Song CH, Ahn CB. The Effects of Decreasement of Blood Pressure on Hypertension Patients by Moxibustion. Journal of Acupuncture Research 2001;18(5):70–6.
15. Han CH, Han CH, Shin MS, Shin SH, Choi SM. The Antihypertensive Effect of Acupuncture Treatment (Gyeok-Pal Sang-Saeng-Yeok-Chim) in Hypertension Patients; Control Study. Journal of Acupuncture Research 2008;25(6):13–22.
16. Lim KT, Hwang EH, Kim BJ, Park IH, Heo I. Chuna Manual Therapy for Essential Hypertension: A Systematic Review. The Journal of Korea CHUNA manual medicine for spine & nerves 2017;12(1):29–42.
17. Cho HJ, Bae DR, Kim HN, Hwang EH. Baduanjin for Hypertension: Systematic Review. The Journal of Korea CHUNA manual medicine for spine & nerves 2017;12(1):43–56.
18. Han IS, Shin JH, Cho YY, Park HM, Yang CS, Jeong MJ, Jang IS, Sun SH. Qu-Ji-Du -Huang-Wan for Essential Hypertension: Systemic Review and Meta-analysis. Korean Journal of Oriental Physiology & Pathology 2017;31(2):118–25.
19. Lee HY, Kang KW, Lee E, Lee SH, Han CH, Jang IS. A Systematic Review on Antihypertensive Effects of Oryeong-san. The Journal of Korean Oriental Internal Medicine 2013;34(3):289–97.
20. Kang KW, Kang JY, Jeong MJ, Kim HJ, Sun SH, Jang IS. The Effect of Cheonmagudeung-eum for Hypertension: A Systematic Review and Meta-analysis. The Journal of Korean Oriental Internal Medicine 2018;39(1):22–43.
21. Abdi H, Tayefi M, Moallem SR, Zhao B, Fayaz M, Ardabili HM, Razavi A, Darbandi M, Darbandi S, Abbasi P, Ferns GA, Majid G. Abdominal and auricular acupuncture reduces blood pressure in hypertensive patients. Complement Ther Med 2017;31:20–6.
22. Cheng L, Li P, Tjen-A-Looi SC, Longhurst JC. What do we understand from clinical and mechanistic studies on acupuncture treatment for hypertension? Chinese Medicine 2015;10(1):36.
23. Chen H, Shen FE, Tan XD, Jiang WB, Gu YH. Efficacy and Safety of Acupuncture for Essential Hypertension: A Meta-Analysis. Medical Science Monitor 2018;24:2946–69.
24. Zhao XF, Hu HT, Li JS, Shang HC, Zheng HZ, Niu JF, Shi XM, Wang S. Is Acupuncture Effective for Hypertension? A Systematic Review and Meta-Analysis. PLOS ONE 2015;10(7):127019.
25. Ministry of Health & Welfare. Study on the introduction of Korean Traditional Medicine Management System 2012.
26. National Health Insurance Service. National Health Insurance Sharing Service Sample research DB; Wonju, Korea: Accessed October 29, 2018. https://nhiss.nhis.or.kr/bd/ab/bdaba002cv.do .
27. Kim JH, Park EC, Kim TH, Lee YH. Hospital Charges and Continuity of Care for Outpatients with Hypertension in South Korea: A Nationwide Population-Based Cohort Study from 2002 to 2013. Korean J Fam Med 2017;38(5):242–8.
28. Lee CJ, Ryu JI, Kim HC, Ryu DR, Ihm SH, Kim YJ, Shin JH, Pyun WB, Kang HS, Park JH, Hwang JS, Park SH. Clinical Benefit of Treatment of Stage-1, Low-Risk Hypertension: Korean Nationl Health Insurance Daabase Analysis. HYPERTENSION 2018;72(6):1285–93.
29. Commissioner of Statistics Korea (KOSTAT). Korean Standard Classification of Diseases 7th editionth ed. Accessed October 15, 2018. https://kssc.kostat.go.kr:8443/ksscNew_web/index.jsp# .
30. Austin PC, Grootendorst P, Anderson GM. A comparison of the ability of different propensity score models to balance measured variables between treated and untreated subjects: a Monte Carlo study. Statistic in Medicine 2007;26(4):734–53.
31. Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, Saunders LD, Beck CA, Feasby TE, Ghali WA. Coding Algorithms for Defining Comorbidities in ICD-9-CM and ICD-10 Administrative Data. Medical Care 2005;43(11):1130–9.
32. Lee HW, Jung MS, Lee KN. A Study on the Patient’s Attitude of Korean Medicine by Social Classes. Journal of Society of Preventive Korean Medicine 2007;11(2):71–86.
33. D’Hoore W, Sicotte C, Tilquin C. Risk adjustment in outcome assessment: The Charlson comorbidity index. Methods of Information in Medicine 1993;32(5):382–7.
34. Haviland A, Nagin DS, Rosenbaum PR. Combining propensity score matching and group-based trajectory analysis in an observational study. PSYCHOLOGICAL METHODS 2007;12(3):247–314.
35. Austin PC. The use of propensity score methods with survival or time-to-event outcomes: reporting measures of effect similar to those used in randomized experiments. Statistics in Medicine 2014;33(7):1242–58.
36. Lee DK. An introduction to propensity score matching methods. Anesthesia and Pain Medicine 2016;11(2):130–48.
37. Heinze G1, Jüni P. An overview of the objectives of and the approaches to propensity score analyses. European Heart Journal 2011;32(14):1704–8.
38. Leem JT. Does acupuncture reduce the risk of acute myocardial infarction? Integrative medicine research: IMR 2016;5(2):165–8.
39. Longhurst J. Acupuncture’s Cardiovascular Actions: A Mechanistic Perspective. Med Acupunct 2013;Apr. 25(2):101–13.
40. Shao M, Li Y, Cui H, Jiang M, Tan Q. Protective effect of acupuncture preconditioning on oxidative stress injury induced by myocardial ischemia-reperfusion injury in rats. Zhongguo Zhen Jiu 2017;Mar. 12. 37(3):285–90.
41. Fu SP, He SY, Xu B, Hu CJ, Lu SF, Shen WX, et al. Acupuncture promotes angiogenesis after myocardial ischemia through H3K9 acetylation regulation at VEGF gene. PLoS One 2014;9(4):e94604.
42. Shiflett SC. Does acupuncture work for stroke rehabilitation: what do recent clinical trials really show? Top Stroke Rehabil 2007;14(4):40–58.
43. Shih CC, Liao CC, Sun MF, Su YC, Wen CP, Morisky DE, Sung FC, Hsu CY, Lin JG. A Retrospective Cohort Study Comparing Stroke Recurrence Rate in Ischemic Stroke Patients with and Without Acupuncture Treatment. MEDICINE 2015;94(39):1572.
44. Shih CC, Yeh CC, Hu CJ, Lane HL, Tsai CC, Chen TL, Liao CC. Risk of dementia in patients with non-haemorrhagic stroke receiving acupuncture treatment: a nationwide matched cohort study from Taiwan’s National Health Insurance Research Database. BMJ Open 2017;7(6):013638.
45. Weng SW, Liao CC, Yeh CC, Chen TL, Lane HL, Lin JG, Shih CC. Risk of epilepsy in stroke patients receiving acupuncture treatment: a nationwide retrospective matched-cohort study. BMJ Open 2016;6(7):010539.
46. Chang CC, Chen TL, Lin CS, Chung CL, Yeh CC, Hu CJ, Lane HL, Liao CC, Shih CC. Decreased risk of pneumonia in stroke patients receiving acupuncture: A nationwide matched-pair retrospective cohort study. PLOS ONE 2018;13(5):0196094.
47. Fang XM, Chen RL, Tang YP, Wang Q, Fang X, Gu GL, Hang SY, Chen SR, Lee CU. Controlled Clinical Research on the Treatment of Hypertension by Tianmagouteng Decoction in Combination with Western Medicine. Guangxi Zhongyiyao Daxue Xuebao 2014;17(4):1–4.
48. Zhong LQ. Refractory hypertension randomized controlled study Xuefuzhuyu Tang combined western medicine. Journal of practical traditional Chinese internal medicine 2015;29(8):127–8.
49. Choi MK, Lee JH, Lee TR, Rhee HS, Rhim KH. The Recognition and Attitude of Users Using the Combined Western-Oriental Medical Institutions. Journal of the Korean Society of Health Information and Health Statistics 2005;30(1):35–44.
50. Park YS, Kim JS. Analysis of the current status outpatient utilization of medical and Korean traditional medical institutions: focuse on outpatient frequent outpatient diseases. Korean Medical Association 2017;60(11):912–919.

Article information Continued

Fig. 1

Flowchart of sample selection process. From 1 million sampled data, patients who were newly diagnosed with hypertension, between 2003 and 2006, were included in this study. Except for subjects that did not meet the criteria, the korean medicine hospital utilization group and the non-korean medicine hospital utilization group group matched with propensity score at a 1:1 ratio.

Fig. 2

Kaplan-Meier Curves on MACE, all-cause mortality, myocardial infarction, and stroke in Korean medicine hospital utilization group and non-Korean medicine hospital utilization group before PSM, after PSM and after stabilized IPTW. MACE indicates Major Adverse Cardiovascular Events; PSM, Propensity Score Matching; and s IPTW, Stabilized Inverse Probability of Treatment Weighting.

Fig. 3

Kaplan-Meier curves on cerebro-cardio vascular mortality in Korean medicine hospital utilization group and non-Korean medicine hospital utilization group before PSM, after PSM and after stabilized IPTW. CSD indicates Circulatory System Disease ; IHD, Ischemic Heart Disease; PSM, Propensity Score Matching; and s IPTW, stabilized Inverse Probability of Treatment Weighting.

Table 1

Baseline characteristics of subjects

Before PSM After PSM After stabilized IPTW

Demographics non-korean medicine hospital utilization korean medicine hospital utilization absolute standardized difference non-korean medicine hospital utilization korean medicine hospital utilization absolute standardized difference non-korean medicine hospital utilization korean medicine hospital utilization absolute standardized difference
N 40798 18328 18242 18242 40832 18311

Age (y) 57.5±10.3 59.2±10.2 0.1729 59.1±10.2 59.2±10.2 0.0116 58.0±10.4 58.0±10.2 0.0055
 40–49 11007 (27.0) 3842 (21.0) 0.1413 3927 (21.5) 3838 (21.0) 0.0119 10321 (25.3) 4637 (25.3) 0.0011
 50–59 11914 (29.2) 5223 (28.5) 0.0156 5173 (28.4) 5207 (28.5) 0.0041 11653 (28.5) 5454 (29.8) 0.0274
 60–69 11755 (28.8) 5816 (31.7) 0.0636 5753 (31.5) 5786 (31.7) 0.0039 12114 (29.7) 5386 (29.4) 0.0056
 70–79 6122 (15.0) 3447 (18.8) 0.1016 3389 (18.6) 3411 (18.7) 0.0031 6744 (16.5) 2835 (15.5) 0.0282

Female sex 18829 (46.2) 11754 (64.1) 0.3676 11631 (63.8) 11674 (64.0) 0.0049 21099 (51.7) 9426 (51.5) 0.0039

Male sex 21969 (53.8) 6574 (35.9) 0.3676 6611 (36.2) 6568 (36.0) 0.0049 19733 (48.3) 8885 (48.5) 0.0039

Household income
 Low 11696 (28.7) 4981 (27.2) 0.0332 4899 (26.9) 4967 (27.2) 0.0084 11501 (28.2) 5139 (28.1) 0.0023
 Middle 13857 (34.0) 6181 (33.7) 0.0051 6199 (34.0) 6160 (33.8) 0.0045 13857 (33.9) 6234 (34.0) 0.0023
 High 15245 (37.4) 7166 (39.1) 0.0356 7144 (39.2) 7115 (39.0) 0.0033 15475 (37.9) 6938 (37.9) 0.0002

Residence Urban 34021 (83.4) 15300 (83.5) 0.0024 15283 (83.8) 15230 (83.5) 0.0079 34065 (83.4) 15284(83.5) 0.0010

Residence Rural 6777 (16.6) 3028 (16.5) 0.0024 2959 (16.2) 3012 (16.5) 0.0079 6767 (16.6) 3028 (16.5) 0.0010

CCI 2.5±1.9 3.1±2.1 0.3074 3.0±2.1 3.0±2.1 0.0007 2.7±2.1 2.7±1.9 0.0008
 0 5494 (13.5) 1454 (7.9) 0.1797 14362 (8.0) 1454 (8.2) 0.0016 4936 (12.1) 2016 (11.0) 0.0337
 1 8383 (20.5) 2805 (15.3) 0.1370 2820 (15.5) 2805 (15.4) 0.0023 7787(19.1) 3404 (18.6) 0.0123
 2 9456 (23.2) 3779 (20.6) 0.0619 3813 (20.9) 3779 (20.7) 0.0046 9103 (22.3) 4125 (22.5) 0.0056
 3 7768 (19.0) 3747 (20.4) 0.0353 3772 (20.7) 3747 (20.5) 0.0034 7864 (19.3) 3652 (19.9) 0.0172
 4 4600 (11.3) 2692 (14.7) 0.1017 2690 (14.7) 2692 (14.8) 0.0003 4898 (12.0) 2332 (12.7) 0.0224
 5 2455 (6.0) 1694 (9.2) 0.1217 1654 (9.1) 1689 (9.3) 0.0067 2767 (6.8) 1331 (7.3) 0.0192
 6 1279 (3.1) 1009 (5.5) 0.1168 928 (5.1) 980 (5.4) 0.0128 1531 (3.7) 735 (4.0) 0.0137
 7 616 (1.5) 530 (2.9) 0.0943 459 (2.5) 499 (2.7) 0.0137 794 (1.9) 357 (1.9) 0.0004
 8 341 (0.8) 298 (1.6) 0.0717 260 (1.4) 293 (1.6) 0.0148 469 (1.1) 185 (1.0) 0.0134
 9 186 (0.5) 157 (0.9) 0.0496 168 (0.9) 156 (0.9) 0.0070 286 (0.7) 91 (0.5) 0.0264
 ≥10 220 (0.5) 160 (0.9) 0.0416 216 (1.2) 148 (0.8) 0.0375 397 (1.0) 83 (0.5) 0.0617

IPTW indicates Inverse Probability of Treatment Weighting; and PSM, Propensity Score Matching.

*

Data are expressed as the mean ± standard deviation, or number (%).

Table 2

Incidence rate of MACE, all-cause mortality, myocardial infarction, and stroke by acupuncture

Before PSM After PSM After stabilized IPTW

Events (n) Follow-up duration (person years) Incidence rate (per 1000 person years) Events (n) Follow-up duration (person years) Incidence rate (per 1000 person years) Events (n) Follow-up duration (person years) Incidence rate (per 1000 person years)
MACE
 non-korean medicine hospital utilization 8921 408,268 21.85 4225 181,387 23.29 9153 407,385 22.47
 korean medicine hospital utilization 3542 180,099 19.67 3521 179,301 19.64 3401 180,666 18.83

All-cause mortality
 non-korean medicine hospital utilization 6145 426,213 14.42 2938 189,896 15.47 6372 425,551 14.97
 korean medicine hospital utilization 2209 188,369 11.73 2197 187,527 11.72 2104 188,577 11.16

Myocardial infarction
 non-korean medicine hospital utilization 1561 419,244 3.72 744 186,559 3.99 1585 418,522 3.79
 korean medicine hospital utilization 695 185,274 3.75 690 184,450 3.74 671 185,580 3.62

Stroke
 non-korean medicine hospital utilization 3098 414,448 7.48 1457 184,355 7.90 3159 413,605 7.64
 korean medicine hospital utilization 1402 182,871 7.67 1395 182,055 7.66 1328 183,364 7.24

IPTW indicates Inverse Probability of Treatment Weighting; MACE, Major Adverse Cardiovascular Events; and PSM, Propensity Score Matching.

Table 3

Hazard ratios of MACE, all-cause mortality, myocardial infarction and stroke treated with acupuncture

Crude analysis Multivariable analysis PSM stabilized IPTW

HR (95% CI) P-value HR* (95% CI) P-value HR (95% CI) P-value HR (95% CI) P-value
MACE 0.91 (0.87–0.94) <.0001 0.84 (0.81–0.88) <.0001 0.84 (0.81–0.87) <.0001 0.84 (0.81–0.88) <.0001
All-cause mortality 0.82 (0.78–0.86) <.0001 0.75 (0.71–0.78) <.0001 0.75 (0.72–0.79) <.0001 0.75 (0.72–0.79) <.0001
Myocardial infarction 1.00 (0.92–1.10) 0.9676 0.96 (0.87–1.05) 0.3380 0.90 (0.83–0.97) 0.0076 0.95 (0.87–1.05) 0.3180
Stroke 1.03 (0.97–1.10) 0.3766 0.96 (0.90–1.02) 0.1825 0.98 (0.92–1.03) 0.3860 0.95 (0.89–1.02) 0.1471

CI indicates Confidence Interval; HR, Hazard Ratio; IPTW, Stabilized Inverse Probability of Treatment Weighting; MACE, Major Adverse Cardiovascular Events; and PSM, Propensity Score Matching.

*

Adjusted for variables listed in the table 1.

Baseline characteristics of subjects

Cox proportional hazards model stratifying on matched pairs.

Weighted Cox proportional hazards model with robust standard errors.

Table 4

Incidence rate of cerebro-cardiovascular mortality by acupuncture

Before PSM After PSM stabilized IPTW

Events (n) Follow-up duration (person years) Incidence rate (per1000 person years) Events (n) Follow-up duration (person years) Incidence rate (per 1000 person years) Events (n) Follow-up duration (person years) Incidence rate (per 1000 person years)
Stroke-related mortality
 non-korean medicine hospital utilization 670 426,213 1.57 333 189,896 1.75 704 425,551 1.65
korean medicine hospital  utilization 263 188,369 1.40 263 187,527 1.40 241 188,577 1.28

Hemorrhage stroke-related mortality
 non-korean medicine hospital utilization 163 426,213 0.38 90 189,896 0.47 170 425,551 0.40
 korean medicine hospital utilization 59 188,369 0.31 59 187,527 0.31 52 188,577 0.28

Ischemic stroke-related mortality
 non-korean medicine hospital utilization 193 426,213 0.45 93 189,896 0.49 204 425,551 0.48
 korean medicine hospital utilization 78 188,369 0.41 78 187,527 0.42 72 188,577 0.38

IHD-related mortality
 non-korean medicine hospital utilization 366 426,213 0.86 167 189,896 0.88 379 425,551 0.89
 korean medicine hospital utilization 139 188,369 0.74 139 187,527 0.74 135 188,577 0.72

CSD-related mortality
 non-korean medicine hospital utilization 1465 426,213 3.44 730 189,896 3.84 1541 425,551 3.62
 korean medicine hospital utilization 580 188,369 3.08 578 187,527 3.08 539 188,577 2.86

CSD indicates Circulatory System Disease ; IHD, Ischemic Heart Disease; IPTW, Inverse Probability of Treatment Weighting; and PSM, Propensity Score Match

Table 5

Hazard ratios of cerebro-cardiovascular mortality by acupuncture

Crude analysis Multivariable analysis PSM stabilized IPTW
HR (95% CI) P-value HR* (95% CI) P-value HR(95% CI) P-value HR(95% CI) P-value
Stroke-related mortality 0.90 (0.78–1.04) 0.1400 0.78 (0.67–0.90) 0.0009 0.80 (0.71–0.91) 0.0008 0.78 (0.67–0.90) 0.0010
Hemorrhage stroke-related mortality 0.83 (0.62–1.12) 0.2195 0.72 (0.53–0.97) 0.0325 0.72 (0.56–0.92) 0.0097 0.70 (0.52–0.95) 0.0236
Ischemic stroke-related mortality 0.93 (0.71–1.20) 0.5629 0.80 (0.61–1.04) 0.0988 0.79 (0.61–1.01) 0.0622 0.80 (0.61–1.05) 0.1146
IHD-related mortality 0.87 (0.72–1.06) 0.1655 0.79 (0.65–0.97) 0.0225 0.79 (0.66–0.94) 0.0097 0.81 (0.66–1.00) 0.0474
IHD-related mortality 0.87 (0.72–1.06) 0.1655 0.79 (0.65–0.97) 0.0225 0.79 (0.66–0.94) 0.0097 0.81 (0.66–1.00) 0.0474
CSD-related mortality 0.91 (0.83–1.00) 0.0508 0.79 (0.71–0.87) <.0001 0.79 (0.72–0.86) <.0001 0.80 (0.72–0.88) <.0001

CI indicates Confidence interval; CSD, Circulatory System Disease, HR, Hazard Ratio; IHD, Ischemic Heart Disease; IPTW, Inverse Probability of Treatment Weighting; and PSM, Propensity Score Matching.

*

Adjusted for variables listed in the table 1.

Cox proportional hazards model stratifying on matched pairs.

Weighted Cox proportional hazards model with robust standard errors.