Reduced Fine Particulate Matter Air Pollution Exposures Using In-Home Portable Air Cleaners: PILOT RESULTS OF THE CARDIAC REHABILITATION AIR FILTER TRIAL (CRAFT) : Journal of Cardiopulmonary Rehabilitation and Prevention

Secondary Logo

Journal Logo

Brief Reports

Reduced Fine Particulate Matter Air Pollution Exposures Using In-Home Portable Air Cleaners


Bard, Robert L. MS; Rubenfire, Melvyn MD; Fink, Samantha BS; Bryant, Joseph BS; Wang, Lu PhD; Speth, Kelly MS; Zhou, Nina MS; Morishita, Masako PhD; Brook, Robert D. MD

Author Information
Journal of Cardiopulmonary Rehabilitation and Prevention 40(4):p 276-279, July 2020. | DOI: 10.1097/HCR.0000000000000516
  • Free

Fine particulate matter <2.5 μm (PM2.5) air pollution is a leading contributor to global morbidity and mortality.1 It accounts for 8.9 million deaths/yr (213 000 in North America alone), with >50% due to cardiovascular diseases.1,2 Exposures increase the risks for myocardial infarctions, strokes, heart failure, and sudden death.2–4 Many potentially responsible mechanisms have been shown including vascular dysfunction, elevated blood pressure, enhanced thrombosis, and arrhythmia, as well as increased atherosclerosis.2,5,6 As such, both the American Heart Association and the European Society of Cardiology recognize PM2.5 as a cardiovascular risk factor.5,6

Reductions in ambient air pollutants decrease morbidity and mortality.7 Unfortunately, mounting studies confirm that low levels of PM2.5 within current United States (US) National Ambient Air Quality Standards (<12 μg·m−3) still pose significant public health risks.2–4 Additional evidence also demonstrates that patients with cardiovascular diseases face the greatest threat.2 As such, we and others have made calls for clinicians to take actions to help protect their patients.8,9 The problem is that there is a paucity of evidence showing what strategies are effective or worthwhile.7 We recently demonstrated in elderly adults living in a senior facility in Detroit that in-home portable air cleaners (PACs) can significantly decrease personal PM2.5 exposures by 40-50%.10 This resulted in a lowering of systolic blood pressures by 3.2 mm Hg within 3 d.10 However, no study has investigated whether PACs remain effective for patients residing in their own homes who are less restricted in daily routines. Before governmental or medical agencies can recommend specific interventions beyond precautionary advice (eg, physical activity restrictions),9 further data are required. We therefore designed the Cardiac Rehabilitation Air Filter Trial (CRAFT) to evaluate the efficacy of PAC usage in patients with heart disease living at home.


The pilot phase of CRAFT was designed to show the feasibility and potential effectiveness of using in-home PACs by 20 patients attending outpatient cardiac rehabilitation (CR) at Michigan Medicine. Exclusion criteria included smoking by the patient or any member of the patient household as determined by self-report and medical history review. Patients were instructed to continue their normal routine provided they sleep in the home during the study. Patients were randomized into a double-blind crossover study of 5 d of active versus sham intervention with a 1-wk washout period between limbs. For the active intervention, PACs had HEPA filters positioned inside the device by a study investigator; the filters were removed for the sham mode. Other study investigators and the patient were blinded to the condition; it was not possible for patients to identify any differences between modes. The PACs were HAP8650B-NU-1 devices (Holmes) for large rooms (310 ft2), which were purchased by investigators at a retailer costing roughly US $120 each. When applicable, the filters were true HEPA (HAPF600TCS) that remove 99.97% of airborne particles at 0.3 μm.7,10 During a routine CR visit, patients were provided with 2 PACs, both set to the same randomized mode by an investigator who maintained the blinding to ensure all other investigators and patients remained blinded to the PAC condition. Patients were instructed to run both PACs in their home continuously for 5 d, one PAC in their bedroom and the other in a main living space. Four days later, after a return CR visit, patients were given a personal PM2.5 monitor, pDR-1500 (Thermo Scientific)7 to wear on a belt or have near their person (eg, bedside) for the following 24 hr. The device recorded 10-min long PM2.5 exposure averages. The lower limit of detection for PM2.5 values measured by the monitor was set at 0.5 μg·m−3. Patients kept a log of their activity every 30 min in an attempt to explain pollution monitoring results, such as driving. After a 1-wk washout period, patients repeated the procedures by crossing over to the remaining PAC condition to complete the protocol. The study was approved by the institutional review board of the University of Michigan, and all participants signed a comprehensive informed consent form.

Data are presented as mean ± SD. Differences in PM2.5 exposure levels between study limbs were tested by the Wilcoxon signed rank test and by linear mixed models with subject-level random intercept to account for correlation within subjects. Analyses were also duplicated after removing subjects with outlying or implausible data such as negative PM2.5 values or >300 μg·m−3.


Eighteen patients completed both study limbs; one patient discontinued the study after one exposure, and one patient missed exposure data because he left the monitor at the clinic. The study was representative of a CR population; they were elderly with cardiovascular disease (Table). Seven patients were in phase 3, whereas 13 were in phase 2 CR. All lived near the facility (mean = 10.7 ± 6.0 mi; maximum: 25.4 mi away) in southeast Michigan, which typically has good air quality. The mean exposure to PM2.5 was 7.1 μg·m−3 during sham intervention (excluding patients 6 and 14 with anomalous high readings).

Table - Patient Characteristics (N = 20)a
Age, yr 70.8 ± 9.6
Female 4 (20)
White 19 (95)
Asian 1 (5)
Health parameters
Body mass index, kg·m−2 29.2 ± 5.4
Systolic blood pressure, mm Hg 118 ± 13
Diastolic blood pressure, mm Hg 65 ± 10
Cardiovascular risk factors
Hyperlipidemia 18 (90)
Hypertension 12 (60)
Diabetes mellitus 7 (35)
Heart failure 8 (40)
Prior cardiovascular event
Myocardial infarction 9 (45)
Percutaneous coronary intervention 9 (45)
Coronary artery bypass surgery 1 (5)
Stroke 1 (5)
Antihypertensive 20 (100)
Statin 19 (95)
aData are presented as mean ± SD or n (%).

Average 24-hr personal-level PM2.5 concentrations were significantly lower during active versus sham (3.6 ± 15.4 vs 14.2 ± 68.8 μg·m−3; P < .001) PAC intervention. Because of anomalous high short-term PM2.5 values (eg, possible exposures to secondhand smoke or cooking), we performed sensitivity analyses removing outlying patients. Outliers did not influence the main results, and the model remained significant after removing 1, 2, 3, and 4 patients with the most variable PM2.5 values. Active PAC intervention significantly reduced PM2.5 exposures by 12.2 μg·m−3 (95% CI, −24.2 to −0.2; P = .047), which was a 44% reduction compared with sham. Exposures were reduced in 16 of the 18 individual patients (Figure).

Waterfall plot of the absolute changes in 24-hr personal exposures to PM2.5 (in μg·m−3) for each participant who completed both study limbs (n = 18) provided by usage of active portable air cleaners versus sham intervention.


We show here for the first time that providing cardiac patients with PACs for use in their homes results in an approximately 40% reduction in personal-level exposure to PM2.5 air pollution. The results are similar to our prior study in a senior facility.10 However, the benefits occurred in free-living patients residing in their own homes in a region with good air quality. These pilot study findings are important because they validate the effectiveness of PACs in a real-world clinical setting and represent a critical next step required to rationalize and design large-scale outcome trials.8,9

Air pollution is a leading cause of global morbidity and mortality.1 In countries where air quality is generally good, such as the US, PM2.5 still causes tens of thousands of deaths per year. More than half of these deaths are from cardiovascular events2–4 because PM2.5 is linearly related to death and disability even at low PM2.5 concentrations (2-12 μg·m−3). Therefore, it is crucial that high-risk individuals in the US, such as CR patients, have access to inexpensive, proven methods to reduce air pollution exposures and thus cardiac risk. The US Environmental Protection Agency currently provides guidance for activity restrictions based on ambient air pollution levels ( However, they represent precautionary advice prudent for large populations. There are no formal recommendations for any personal-protective strategy (eg, PAC) due to a paucity of supporting data. As such, we and others have called for more research to help governmental and medical agencies formulate evidence-based strategies to reduce exposures.7–9 We also believe that any intervention should be tested in outcome trials to determine health benefits.

It remains speculative whether PACs prevent cardiovascular events. Prior studies have shown improvements in surrogate endpoints including blood pressure, endothelial function, inflammation and oxidative stress, and metabolic health.7,10 However, they were conducted in cities with high levels of air pollution (eg, Shanghai) or among individuals with reduced activity (eg, senior facilities, dormitories). For PACs to be widely effective, their benefits must also apply to free-living patients—as we herein show. Our positive results are likely related to the average Americans spending the majority of time inside their own home, including 6-9 hr/d in their bedroom.7 Assuming PACs are 40% effective and used in typical urban/suburban US populations,1–4 this should decrease PM2.5 exposures by 5 μg·m−3. Conservative estimates are that myocardial infarctions, strokes, and heart failure will decrease by 5% over a few years.2 PACs are safe and inexpensive (∼US $120) and can be used for years with minimal costs (US $10-$30 for replacement HEPA filters and electricity). PACs may be prescribed to millions of patients and thereby reap substantial public health benefits. However, the relative risk for events is larger in patients with cardiovascular disease than in the general population.11,12 We believe this exposure reduction would translate into an approximately 20% decrease in ischemic events if applied to cardiac patients. Finally, PAC use in countries with high levels of air pollution, such as China, will reduce exposure by 40-50 μg·m−3 and thereby yield greater health benefits.7,8

Our results are from a pilot study where the design and sample size were based upon our prior findings.10 Four patients were exposed to extremely high PM2.5 levels (>300 μg·m−3) for brief periods of time (10-30 min), which is implausible for ambient air pollution in Michigan. These data must represent point sources such as active smoking, secondhand smoke, wood/leaf burning, and/or indoor cooking.


In-home PACs can lower exposure to PM2.5 air pollution among cardiac patients living in a stereotypical American home setting. These findings support moving forward with the full-scale CRAFT that aims to show that PACs improve surrogate health endpoints in cardiac patients. Such data are required to design future outcome trials that are urgently needed to help inform on how to protect the global public health from the threats posed by air pollution.


This study was funded by the Michigan Medicine Frankel Cardiovascular Center inaugural grant program.


1. Burnett R, Chen H, Szyszkowicz M, et al. Global estimates of mortality associated with long-term exposure to outdoor fine particulate matter. Proc Natl Acad Sci U S A. 2018;115(38):9592–9597.
2. Rajagopalan SR, Al-Kindi SG, Brook RD. Air pollution and cardiovascular disease. J Am Coll Cardiol. 2018;72(17):2054–2070.
3. Liu C, Chen R, Sera F, Vicedo-Cabrera AM, et al. Ambient particulate air pollution and daily mortality in 652 cities. N Engl J Med. 2019;381:705–715.
4. Di Q, Wang Y, Zanobetti A, et al. Air pollution and mortality in the Medicare population. N Engl J Med. 2017;376(26):2513–2522.
5. Brook RD, Rajagopalan S, Pope CA III, et al. Particulate matter air pollution and cardiovascular disease. An update to the scientific statement from the American Heart Association. Circulation. 2010;121(21):2331–2378.
6. Newby DE, Mannucci PM, Tell GS, et al. Expert position paper on air pollution and cardiovascular disease. Eur Heart J. 2015;36(2):83b–93b.
7. Bard RL, Ijaz MK, Zhang JJ, et al. Interventions to reduce personal exposures to air pollution: a primer for health care providers. Global Heart. 2019;14(1):47–60.
8. Brook RD, Newby DE, Rajagopalan S. The global threat of outdoor ambient air pollution to cardiovascular health. Time for intervention. JAMA Cardiol. 2017;2(4):353–354.
9. Hadley MB, Baumgartner J, Vedanthan R. Developing a clinical approach to air pollution and cardiovascular health. Circulation. 2018;137(7):725–742.
10. Morishita M, Brook RD. Letter regarding: portable air filtration systems, personal-level exposure to fine particulate matter, and blood pressure levels among residents in a low-income senior facility in Detroit. JAMA Intern Med. 2019;179(2):275–276.
11. Chen H, Burnett RT, Copes R, et al. Ambient fine particulate matter and mortality among survivors of myocardial infarction: population-based cohort study. Environ Health Perspect. 2016;124(9):1421–1428.
12. Tibuakuu M, Michos ED, Navas-Acien A, Jones MR. Air pollution and cardiovascular disease: a focus on vulnerable populations worldwide. Curr Epidemiol Rep. 2018;5(4):370–378.

air pollution; cardiac rehabilitation; cardiovascular risk; environment

© 2020 Wolters Kluwer Health, Inc. All rights reserved.