Figure 1
Description of the articles selection processes.
Table 1
Description of included articles on CVD morbidity/mortality and environmental factors.
| AUTHOR & YEAR | LOCATION | STUDY DESIGN | MAIN EXPOSURE(S) | OUTCOME AND DEFINITION | MAIN FINDINGS | STUDY QUALITY |
|---|---|---|---|---|---|---|
| Buadong et al., 2009 (33) | Bangkok, Thailand | Time-series | PM10, O3 | Morbidity – daily hospital visits | There was no significant association for either PM10 or O3 on CVD morbidity in the 3-day cumulative lag model. | Fair |
| Dong et al., 2013 (56) | Liaoning Province, China | Cross-sectional | PM10, SO2, NO2, O3 | Morbidity – Positive response from questionnaire | No significant association was found between any of the air pollutants and CVD morbidity | Fair |
| Tong et al., 2014 (36) | Tianjin Municipality, China | Time-series | PM10, SO2, NO2 | Morbidity – Database | A 10 µg/m3 increase in the 2-day average concentration of PM10 and SO2 were associated with a 0.19% (0.08–0.31) and 0.43% (0.03–0.84) increase in CVD morbidity respectively. No significant association was found for NO2. | Fair |
| Giang et al., 2014 (31) | Thai Nguyen, Vietnam | Time-series | Temperature | Morbidity – Hospital admission | Over a 0–30-day lag period, there was a 12% (1%–25%) increase in CVD hospital admissions per 1 degree below the temperature threshold. A positive, yet non-significant association was observed for increased temperature. | Fair |
| Su et al, 2016 (34) | Haidian District, Beijing, China | Time-series | PM10, PM2.5, SO2, NO2 | Morbidity – Medical records of emergency visits | In the 0–7-day cumulative lag model, no significant association between PM2.5, PM10, SO2, or NO2, and CVD morbidity was observed. | Fair |
| de Freitas et al., 2016 (35) | Victoria, Brazil | Time-series | PM10, O3, SO2, | Morbidity – Hospital records | In the 0–5-day cumulative lag model, CVD events increased by 2.11% (1.06–3.18) per 10 µg/m3 increase in O3. No significant association was observed for PM10 and SO2. | Poor |
| Phung et al., 2016 (39) | Vietnam | Time-series | PM10, SO2, NO2, O3 | Morbidity – Hospital admission | In the lag-3 model, neither PM10, NO2, SO2 or O3 had a statistically significant association with CVD morbidity. | Fair |
| Ma et al., 2017 (42) | Beijing, China | Time-series | PM10, SO2, NO2 | Morbidity – Hospital admission | For a 10 µg/m3 increase in NO2, ER cardiovascular admission increased by 1.4% (RR:0.986; 95%CI:0.976–0.996) in the 0–6-day cumulative lag model. There was no association between CVD admission and PM10 or SO2. | Fair |
| Liu et al., 2018 (46) | Mainland China | Case crossover | CO | Morbidity – Health database | A 1 mg/m3 increase in the same day CO was associated with a 4.39% (4.07–4.70) increase in CVD. | Fair |
| Li et al., 2018 (47) | Beijing, China | Case crossover | CO | Morbidity – Health database | A 1 mg/m3 increase in the 2-day moving average of CO was associated with a 2.8% (2.2–3.3) increase in daily hospital CVD admissions. | Fair |
| Phosri et al., 2019 (38) | Bangkok, Thailand | Time-series | SO2, NO2, O3, CO | Morbidity – Daily hospital admission | A 10 µg/m3 increase in PM10, SO2, and NO2 corresponded to 0.6% (0.10–1.00), 5.3% (2.42–8.21), and 0.6% (0.06–1.09) increases in total CVD admission in the 0–4-day cumulative lag models, respectively. A 1 mg/m3 increase in CO increased CVD admission by 4.2% (1.35–7.26). No significant association with O3 was observed. | Fair |
| Yao et al., 2019 (86) | Yichang Province, China | Time-series | PM10, PM2.5 | Morbidity-Daily inpatient records | There was no statistically significant association between PM10 or PM2.5 and CVD admission in the lag 7 model. | Fair |
| Amsalu et al., 2019 (32) | Beijing, China | Time-series | PM2.5 | Morbidity – Daily hospital admission | In the 0–3-day lag model, a 10 µg/m3 increase in PM2.5 was associated with a 0.7% (0.4–0.9) increase in CVD hospital admissions. | Fair |
| Cheng et al., 2019 (48) | Lanzhou city, China | Time-series | CO | Morbidity – Daily CVD hospitalization | In the lag 0–4 model, a 1 mg/m3 increase in CO was associated with an 11% increase (95%CI: 3%–20%) in CVD hospitalization. | Fair |
| Khan et al., 2019 (45) | Dhaka, Bangladesh | Case crossover | PM2.5 | Morbidity – Emergency room visit | An IQR increase (103 µg/m) of PM2.5 corresponded to a 15% increase (1–30) in CVD emergency room visits in the 3–5-day lag model. | Fair |
| Phosri et al., 2020 (43) | Bangkok, Thailand | Time-series | Temperature | Morbidity – Daily hospital admission | In the 0–21 lag models, an “extremely high” diurnal temperature range (11.6°C) was associated with a 20.6% (0.2–45.2) increase in CVD hospital admissions. | Fair |
| Rahman et al., 2022 (40) | Dhaka, Bangladesh | Time-series | Temperature | Morbidity – Count of CVD from Database | There was no association between a 1°C increase in temperature variability and ED visits for cardiovascular disease. | Fair |
| Karbakhsh et al., 2022 (44) | Iran | Case crossover | PM10, PM2.5, PMcoarse | Morbidity – CVD admitted | An IQR increase in PMcoarse (IQR: 55 µg/m3) and PM10 (IQR: 71 µg/m3) was associated with an increase in CVD admission (OR:1.02; 95% CI: 1.00–1.05 and 1.02; 95% CI:1.01–1.04) respectively in the lag 0–1–2 model. No significant effect was observed for PM2.5. | Fair |
| Makunyane et al., 2023 (37) | Cape Town, South Africa | Time-series | Temperature | Morbidity – Daily counts of hospital admission | An IQR (6.4°C) increase in temperature variability of TV was associated with a 2.61% (1.15–4.08) increase in hospital admissions. | Fair |
| Ji et al., 2021 (49) | Mainland China | Cohort | Solid fuel | Morbidity – Response from questionnaire | Individuals using solid fuels at baseline had a higher risk of non-fatal CVD event than those using clean fuels (HR:1.18; 95% CI:1.05–1.32). | Fair |
| Liu et al., 2021 (50) | Mainland China | Cohort | PM2.5 | Morbidity – Based on Disease classification | An IQR increase in PM2.5 (27.9 µg/m3) increased the risk of CVD morbidity (HR:1.291, 95% CI: 1.147–1.54). | Fair |
| Mai et al., 2032 (51) | Mainland China | Cohort | PM2.5 | Morbidity – Response from questionnaire | A 10 µg/m3 increase in PM2.5 was associated with an increased risk of CVD morbidity (OR:1.18 95% CI: 1.12–1.26). | Fair |
| Wen et al., 2023 (52) | Mainland China | Cohort | Solid fuel | Morbidity – Self Assessment | Treatment effect of cardiovascular disease after implementation of coal-to-gas/electricity project was not statistically significant. | Fair |
| Wang et al., 2023 (53) | Mainland China | Cohort | NO2 | Morbidity – Questionnaire | A 10 µg/m3 increase in NO2 resulted in an elevated risk of CVD morbidity (HR:1.558 95% CI: 1.477–1.642). | Fair |
| Liu et al., 2023 (54) | Mainland China | Cohort | Solid fuel | Morbidity – Response from questionnaire | The use of solid fuel for cooking and heating versus clean fuel increased the risk of nonfatal CVD incident by 43.0% [HR:1.43 (1.07–1.92)]. | Fair |
| Zhu et al., 2024 (55) | Mainland China, | Cohort | O3 | Morbidity – Questionnaire | A 10 µg/m3 increase in long-term O3 exposure was positively associated with incident of CVD (HR:1.078 95% CI: 1.050–1.106). | Fair |
| Xia et al., 2023 (85) | Mainland China | Cohort | PM2.5 | Morbidity & Mortality – Questionnaire | A 10 µg/m3 increase in PM2.5 was positively associated with total CVD morbidity (HR:1.12, 95% CI: 1.11–1.14) and CVD mortality (HR:1.12 95% CI: 1.08–1.15). | Good |
| Liang et al., 2020 (84) | Mainland China | Cohort | PM2.5 | Morbidity & Mortality – Extracted from questionnaire | A 10 µg/m3 increase in PM2.5 gave HRs for CVD incidence and mortality of 1.25(1.22–1.28) and 1.16 (1.12–1.21), respectively. | Good |
| Jalali et al., 2021 (23) | Isfahan, Iran | Cohort | PM2.5 | Morbidity & Mortality – Questionnaire & health records | The risk of CVD event increased by 2.6% (OR:1.026, 95% CI:1.016–1.036) for a 10 µg/m3 increase in PM2.5. No significant association was observed between PM2.5 and CVD mortality. | Fair |
| Zhang et al., 2006 (57) | Shanghai, China | Time-series | O3 | Mortality – Database | An increase of 10 µg/m3 in the 4-day O3 average corresponded to a 0.9% increase (95% CI: 0.5–1.4) in total cardiovascular mortality. | Fair |
| Tam et al., 2010 (58) | Hong Kong Administrative Region | Time-series | Temperature | Mortality – Database | In the 0–3 lag model, a 1°C increase in diurnal temperature range resulted in a 1.7% increase in cardiovascular mortality (RR:1.017, 95% CI: 1.003–1.033) | Poor |
| Yang et al., 2012 (87) | Suzhou Province, China | Time-series | O3 | Mortality – Database | An IQR increase in the 24-hour average concentration of O3 (33.3 µg/m3) was associated with a 3.33% (95% CI: 0.50–6.16) increase in CVD mortality. | Fair |
| Chen et al., 2012 (60) | Mainland China | Time series | SO2 | Mortality – Database | A 10 µg/m3 increase in the 2-day moving average of SO2 was associated with a 0.83% increase in cardiovascular mortality (95% PI:0.47–1.19). | Fair |
| Wichmann & Voyi, 2012 (76) | South Africa | Case crossover | PM10, SO2, NO2, | Mortality – Database | There was a 3.4% (0.3–6.6) and 2.6% (0.1–5.2) increase in cardiovascular mortality per IQR increase in NO2 (IQR: 12 µ/m3) and SO2 (IQR: 8 µg/m3), respectively. No significant effect of PM10 was observed. | Fair |
| Fuhai Geng et al., 2013 (61) | Shanghai, China | Time-series | BC & PM2.5 | Mortality – Database | An IQR increase in the mean daily concentrations of BC (IQR: 2.7 µg/m3) and PM2.5 (IQR: 41.8 µg/m3) corresponded to a 3.2% (0.6–5.7) and 3.3% (0.4–6.1) increase in total cardiovascular mortality, respectively. | Fair |
| Wang et al., 2014 (62) | Suzhou Province, China | Time-series | Temperature | Mortality – Database | In the 0–28 lag model, extreme cold (1st centile: –0.3°C) and hot (99th centile: 32.6°C) temperatures were positively associated with cardiovascular mortality with RRs of 2.67 (1.64–4.33) and 1.62 (1.21–2.17), respectively. | Fair |
| Han et al., 2017 (63) | Jinan Province, China | Time-series | Temperature | Mortality – Database | Cold spells (3 consecutive days below –3.8°C) and heat waves (3 consecutive days above 29°C) were associated with CVD mortality RRs of 1.06 (1.03–1.10) and 1.03 (1.00–1.06), respectively | Fair |
| Chen et al., 2018 (65) | Mainland China | Time-series | NO2 | Mortality – Database | A 10 µg/m3 increase in the 2-day average concentration of NO2 would increase total cardiovascular mortality by 0.9% (0.7–1.2) | Fair |
| Chen et al., 2018 (64) | Mainland China | Time-series | PM2.5 | Mortality – Database | In the 0–2 lag model, no significant association between PM2.5 and cardiovascular mortality was observed. | Fair |
| Liu et al., 2018 (66) | Mainland China | Time-series | CO | Mortality – Database | In the 0–1 lag model, a 1 mg/ m3 increase in CO was associated with a 1.12% (PI:0.42–1.83) increase in cardiovascular mortality | Fair |
| Wu et al., 2018 (67) | Guangzhou Province, China | Time-series | PM2.5, PM10 & PM10-2.5 | Mortality – Database | In the lag 06 model, a 10 µg/m3 increase in PM2.5, PM coarse, and PM10 was associated with an excess risk for CVD mortality of 1.15% (95% CI: 0.68, 1.62), 1.64% (95% CI: 0.86, 2.43), and 0.82% (95% CI: 0.49, 1.14), respectively. | Fair |
| Zhang et al., 2019 (41) | Jiangsu Province, China | Time-series | O3 | Mortality – Database | In the lag 0–3 model, a 10 µg/m3 increase in O3 was associated with a 0.983% (0.588–1.3770) increase in CVD-related death. | Fair |
| Liu et al., 2019 (68) | Shenyang Province, China | Time-series | PM10, PM2.5, SO2, NO2, O3, CO | Mortality – Death registry | In the lag 05 model, 10 µg/m3 increases in PM2.5, PM10, SO2, and NO2 were associated with RRs for CVD mortality of 1.004 (1.001, 1.008), 1.003 (1.001, 1.006), 1.005 (1.001, 1.009), and 1.016 (1.005, 1.028), respectively. A 1 mg/m3 increase in CO was associated with an RR of 1.066 (1.025, 1.108). No significant association was observed for O3. | Fair |
| Duan et al., 2019 (69) | Shenzhen Province, China | Time-series | NO2 | Mortality – Database | In the lag 0–5-day model, a 10 µg/m3 increase in NO2 was associated with a 3.41% (1.55–5.30) increase in cardiovascular mortality. | Fair |
| Iranpour et al., 2020 (70) | Ahvaz, Iran | Time-series | Temperature | Mortality – Database | In the 0–28-day lag model, no association between heat above the 99th centile (41.2°C) or below the 1st centile (9.3°C), and CVD mortality was observed. | Fair |
| Khosravi et al., 2020 (71) | Mashhad, Iran | Time-series | PM10, PM2.5, NO2, O3, CO | Mortality – Database | None of the five pollutants assessed were associated with cardiovascular mortality. | Fair |
| Zhou et al., 2021 (72) | Taiyuan Province, China | Time-series | PM10, PM2.5 | Mortality – Database | In the 0–30 lag model, a 10 µg/m3 increase in PM2.5 and PM10 was associated with a 3.10% (0.86–5.38) and 1.61% (0.69–2.54) increase in cardiovascular mortality. | Fair |
| Li et al., 2021 (73) | Guangzhou Province, China | Time-series | O3 | Mortality – Registry | In the 0–3 lag model, a 10 µg/m3 increase in O3 was associated with a 0.59% (0.30–0.88) increase in CVD mortality. | Fair |
| Olutola et al., 2023 (75) | South Africa | Case crossover | PM10, SO2, NO2, | Mortality – Database | In the 0–6-day lag model, none of the examined pollutants were associated with increased CVD mortality. | Fair |
| Xia et al., 2023 (74) | Chengdu, China | Time-series | Temperature | Mortality – Database | In the 0–14-day lag model, extreme heat (99th centile, >29 °C) and extreme cold (1st centile, < 3°C) were both associated with increased CVD mortality, with RRs of 1.28 (1.14–1.43) and 1.45 (1.24–1.68), respectively. | Fair |
| Cao et al., 2011 (59) | Mainland China | Cohort | SO2, TSP, NOX | Mortality – Hospital records | A 10 µg/m3 increase in TSP, SO2, and NOx corresponded to 0.9% (95% CI: 0.3, 1.5), 3.2% (95% CI: 2.3, 4.0), and 2.3% (95% CI:0.6, 4.1) increases in cardiovascular mortality, respectively. | Fair |
| Wong et al., 2015 (77) | Hong Kong, Administrative Region | Cohort | PM2.5 | Mortality – Death registry | A 10 µg/m3 increase in PM2.5 exposure was associated with a 22% increase in cardiovascular mortality [HR:1.22 (1.08–1.39)]. | Fair |
| Yu et al., 2018 (78) | Mainland China | Cohort | Solid fuel | Mortality – Questionnaire | Solid fuel use for cooking or heating was significantly associated with higher risk of cardiovascular mortality [HR:1.20 (1.02–1.41)] and [HR:1.29 (1.06–1.55)], respectively. | Fair |
| Yang et al., 2018 (79) | Mainland China | Cohort | PM2.5, NO2 & BC | Mortality – Database | An IQR increase in PM2.5 (5.5 µg/m3) or BC (9.6 µg/m3) was associated with increased HRs for CVD mortality (1.06 [1.02–1.10] and 1.07 [1.02–1.11], respectively. No significant association was observed for NO2. | Fair |
| Arku et al., 2020 (80) | China, India, South Africa and Tanzania | Cohort | Kerosene | Mortality – Hospital records, Death certificate and Verbal autopsies | Household cooking primary with kerosene had a 34% [HR:1.34 (1.08–1.66)] increase in major cardiovascular disease mortality. | Fair |
| Liang et al., 2022 (81) | Mainland China | Cohort | PM2.5 | Mortality – Death registry | A 10 µg/m3 increase in PM2.5 was associated with a HR for cardiovascular mortality of 1.02 (1.00–1.05). | Good |
| Liu et al., 2022 (82) | Yinzhou Province, China | Cohort | O3 | Mortality – Death registry | A 10 µg/m3 increase in long-term annual average of O3 increased cardiovascular mortality by approximately 22% [HR:1.22 (1.12–1.33)]. | Good |
| Niu et al., 2022 (83) | Mainland China | Cohort | O3 | Mortality – Death registry | A 10 µg/m3 increase in O3 was associated with an elevated risk of cardiovascular mortality [HR:1.093 (1.046–1.142)]. | Good |
Figure 2
Meta-analysis of short-term PM2.5 exposure and CVD morbidity and mortality.
Figure 3
Meta-analysis of short-term NO2 exposure and CVD morbidity and mortality.
Figure 4
Meta-analysis of short-term O3 exposure and CVD morbidity and mortality.
Figure 5
Meta-analysis of long-term PM2.5 exposure and cardiovascular disease morbidity and mortality.