Article Text

Download PDFPDF

Indoor air quality in prisons before and after implementation of a smoking ban law
  1. S K Proescholdbell1,
  2. K L Foley2,
  3. J Johnson1,
  4. S H Malek1
  1. 1
    Tobacco Prevention and Control Branch, Division of Public Health, North Carolina Department of Health and Human Services, Raleigh, NC, USA
  2. 2
    Department of Social Sciences and Health Policy, Division of Pubic Health Sciences, Wake Forest University School of Medicine, Winston-Salem, NC, USA
  1. Scott Proescholdbell, Tobacco Prevention and Control Branch, Division of Public Health, North Carolina Department of Health and Human Services, 1932 Mail Service Center, Raleigh, NC 27699–1932, USA; scott.proescholdbell{at}


Objective: To ascertain whether a new indoor smoking ban law in North Carolina correctional facilities was successfully implemented and whether the indoor air quality has improved as a result.

Method: Before the law came into effect, we tested the air quality of 22 dormitory and common areas within six North Carolina prisons using standard protocols for testing particulate matter. We measured particulate matter 2.5 μm in diameter (PM2.5) using state of the art TSI SidePak monitors. After the law went into effect, the same locations within each prison were tested again. Written inmate surveys were also conducted at two prisons, one with partial smoking ban (indoors only) and one with a total smoking ban (indoors and outdoors).

Results: The findings indicate that, on average, levels of respirable suspended particulates (RSPs), an accepted marker for secondhand smoke (SHS) levels, decreased 77% in these prisons after the law took effect compared to levels obtained before ban implementation. Several areas were tobacco-free before the implementation of this ban. In those areas no significant decreases in RSPs were noted.

Conclusion: Laws banning tobacco use in correctional facilities can significantly reduce indoor SHS exposure among inmates, visitors and staff and potentially lead to reduced use. To date, 24 US states have enacted 100% smoke-free correctional facility polices for all indoor areas even though inmates and staff have much higher tobacco use prevalence rates than the general population. With an estimated nine million people incarcerated worldwide, prison smoking bans could have a substantial impact in terms of health outcomes and long-term costs if they can effectively reduce exposure to secondhand smoke.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Recent data have shown that secondhand smoke is a serious health hazard, causing disease and death in non-smokers. A systematic review of these findings was published in 2006 by the US Surgeon General in a comprehensive scientific report, The Health Consequences of Involuntary Exposure to Tobacco Smoke, which indicates that there is no risk-free level of exposure to secondhand smoke (SHS). The report also states that non-smokers exposed to SHS at home or work increase their risk of developing heart disease by 25–30% and lung cancer by 20–30%. These findings are a major public health concern as nearly half of all non-smoking Americans are still regularly exposed to SHS. One of the report’s six conclusions states, “eliminating smoking in indoor spaces fully protects non-smokers from exposure to secondhand smoke. Simply separating smokers from non-smokers, cleaning the air and ventilating buildings cannot eliminate exposures of non-smokers to secondhand smoke.”1

As indicated by the Surgeon General’s report, SHS exposure and its adverse health effects are preventable. This is most effectively done by enacting policies requiring smoke-free facilities.2 As of 1 January 2006, North Carolina, a large southern US state with a population of more than 8.8 million people and 39 000 inmates, required all indoor prison areas to be smoke-free for prisoners, staff and visitors (Session Law 2005–372). This legislation “chips away” at North Carolina’s “Pre-emption” law passed by the General Assembly in 1993 that required state-controlled buildings to set aside 20% of space for smoking, if practicable, and pre-empted stricter local smoking control rules (Smoking in Public Places, GS 143-595-601). In addition, three pilot sites in this study implemented total bans (indoors and outdoors).

The law creating this smoking ban is especially relevant for reducing SHS exposure in correctional facilities. Tobacco use prevalence is exceptionally high among the US incarcerated population with smoking rates estimated to be as high as 70%, well above the 21% rate for all US adults.3 Although several US states have passed tobacco-free policies for correctional facilities, one study demonstrated a lack of effectiveness of the policy stating that 76% of inmates who smoked before a smoking ban continued to do so following the ban using self-reported data.4 Others have also recently questioned the overall effectiveness of smoking bans.5 However, there are no published studies that document indoor air quality before and after such a ban, which provides an unambiguous assessment of the smoking ban and has significant public health implications for SHS exposure. While smoking-cessation counselling and low-cost nicotine-replacement therapy was made available to inmates following the ban on smoking in North Carolina prisons, this report describes indoor air quality at 22 locations within six North Carolina prisons before and after implementation of Session Law 2005–372 requiring all North Carolina correctional facilities to be tobacco free indoors as of 1 January 2006.



North Carolina Department of Correction staff and the research team identified six prisons in North Carolina in which to measure air quality before and after tobacco-free policy implementation. The sites were purposely selected to represent a range of correctional facilities in terms of security level, inmate gender, size and geographic location. The sample consisted of one minimum security women’s prison, one maximum security men’s prison, three medium security men’s prisons and one minimum security men’s prison (table 1). In addition, two sites were selected for in-person, written surveys assessing tobacco use, dependence, desire to quit, agreement with ban and cessation services. One survey site implemented a total ban (indoors and outdoors) while the other implemented a partial ban (indoors only).

Table 1 Prison facility characteristics, 2005 and 2006

Four of six prisons allowed indoor tobacco use before policy implementation. In those that permitted tobacco use, it was allowed in most dormitories and common spaces before the ban. However, they also offered limited designated smoke-free areas in dormitories and staff offices before the ban. Two facilities had already adopted comprehensive indoor tobacco bans at the local level before the statewide ban.

Data collection

A team of state health department personnel, trained in the use of air quality monitoring and surveying, conducted all data collection. Baseline data were collected during November and December 2005 from a sample of 22 locations within six prisons across North Carolina. Follow-up data were collected in all the same six prisons and locations during October-December of 2006, 10–12 months after implementing the ban.

Air quality monitoring in this study measured respirable suspended particles (RSPs) that were 2.5 μm in size, known as particulate matter 2.5, or PM2.5. Particles of this size are released in substantial amounts from burning cigarettes and are easily inhaled deep into the lungs and serve as an accurate proxy for exposure to SHS.6 Air quality monitoring was performed using the TSI SidePak AM510 Personal Aerosol Monitor (TSI, Inc, St Paul, MI, USA). The SidePak uses a built-in sampling pump to draw air through the device, which then measures the real-time concentration in milligrams per cubic metre of PM2.5. The SidePak was zero-calibrated before each use according to manufacturers’ instructions. The SidePak was used with a calibration factor setting of 0.32, suitable for tobacco smoke. Air monitors were calibrated and started outside most facilities to obtain an ambient air quality baseline reading for 5 minutes before entering the main facility. At each prison, air quality data were collected from at least one dormitory and the main lobby or office area. The monitor was placed in a central location on a table, desk or chair in each testing area. Air quality measurements were taken at one-second intervals and averaged over 1 minute.

Observational data were also collected in each location for air monitoring. Data included room dimensions, number of individuals in the room and number of lit cigarettes. The number of people and number of burning cigarettes in each space were recorded every 10 minutes during data collection, and the average number of people and average number of burning cigarettes were calculated. The volume of each location was also measured by estimating room length, width and height, and the cigarette density was calculated by dividing the average number of burning cigarettes by the location volume.

When indoor air quality monitoring was complete at each site, readings were again collected outside each facility for 5 minutes to clearly identify the return to baseline ambient levels in order to accurately interpret measurements. The follow-up measurements were taken on the same day of the week, within a similar timeframe and location as the measurements taken before the smoke-free law was implemented.

The mean baseline time spent in each prison was 77 minutes (range 43–91 minutes) including outside air measurements before and after entering the facility. The mean time spent in each prison at follow-up was 96 minutes (range 51–156 minutes).

In addition, brief written surveys were group administered to inmates at two facilities: one total tobacco ban facility (tobacco-free indoors and on all grounds) and one partial tobacco ban facility (tobacco-free indoors only). Surveys were group administered by research staff. The research staff, with cooperation from correctional facility staff, invited inmates to participate. The research team read consent statements and a 15-item questionnaire to inmates. Participation was voluntary and anonymous and approved by the institutional review board of the North Carolina Department of Correction. A total of 209 inmates participated.

Data analysis

Data analysis for the indoor air quality monitoring proceeded as follows. Firstly, a “location-level” analysis calculated room size and number of burning cigarettes standardised per 100 m3 using direct observation data. The average concentration of particulate matter per 2.5 μm (PM2.5) was also measured for each location using data retrieved from the air-quality monitor. The monitor recorded measurements every minute, which were averaged over time for each facility. The first and last minute of the logged data were removed, and the remaining data points were averaged to provide an average concentration of PM2.5 within each prison location before and after the indoor tobacco ban.

Location data were then combined and reanalysed based on actual and observed policy compliance. Firstly, all data were pooled to evaluate particulate matter concentrations pre-ban and post-ban for all correctional facilities regardless of observed policy compliance (n = 14 sites). One facility that was observed to be out of compliance with the locally imposed tobacco-free ban at baseline was then removed from the analysis and data were reanalysed (n = 13 sites).

For all analyses, the percentage change of PM2.5 levels was determined by comparing the average PM2.5 levels in each prison before the law went into effect with levels after the law was implemented. The Wilcoxon signed rank test was used to assess changes between pre-ban and post-ban PM2.5 levels. All data were analysed using SPSS 14.0 for Windows.7

Survey data were analysed using Stata 7.0. Inconsistent or missing data were excluded from the dataset.8 Data were stratified by type of ban (total or partial) and bivariate analyses were computed on smoking prevalence.


Table 2 shows the average number of cigarettes used and the average PM2.5 concentration by location. The average PM2.5 concentration was substantially lower after the indoor tobacco-use ban went into effect in 12 of 14 locations where indoor smoking and SHS exposure had been observed at baseline. Of the two prisons with higher PM2.5 concentrations at follow-up, one facility had notably higher concentrations (83.25 μg/m3 before and 153.45 μg/m3 after). This occurred at a facility where the smoking ban was clearly not enforced and smoking in the facility buildings was directly observed by the study team despite the new statewide ban.

Table 2 Location-level analysis of the change in concentrations of respirable suspended particles after implementation of a smoking ban, stratified by smoking status—North Carolina, 2005–6

Particulate matter concentration in locations that allowed smoking before tobacco use ban implementation was 93.11 μg/m3 and the average PM2.5 concentration in these locations after the implementation of the tobacco use ban was 21.82 μg/m3 (p<0.006), representing a 77% reduction in SHS exposure. The average PM2.5 concentration in the smoke-free facilities areas before the statewide tobacco-free policy implementation was 11.14 μg/m3 and the average PM2.5 concentration at follow-up measurement was 7.52 μg/m3, a non-significant reduction (table 2).

Figure 1 shows an overall reduction of 77% in exposure to SHS after the ban compared to data before the ban in those prison facilities areas that allowed smoking before 2006 (p<0.006). Excluding one prison not in compliance with the state indoor tobacco ban at follow-up, results demonstrated a 91% reduction in mean PM2.5 concentrations before and after the statewide ban (94.01 μg/m3 to 9.85 μg/m3; p<0.001) (see fig 1).

Figure 1 Particulate matter concentration (PM2.5) overall and among facilities in compliance before and after a statewide indoor tobacco ban in correctional facilities.

Survey results

Overall, 65% of the inmates that responded to this study survey smoked some days or every day. After ban implementation, smoking prevalence was higher in the partial-ban versus the total-ban facility (64% to 42% daily smokers and 20% to 5% non-daily smokers, respectively) p<0.001. One out of three (33%) of those who smoked expressed an interest in a programme to help them quit smoking; there was no difference between facilities. Additionally, 50% of respondents expressed interest in no cost nicotine replacement therapy.


The findings indicate that levels of respirable suspended particulates (RSPs), an accepted marker for exposure to SHS, decreased 77% in North Carolina prisons after the implementation of an indoor smoking ban at six state prisons in those areas that previously allowed smoking before the ban. In prison areas that were already smoke-free at baseline, no significant reductions were noted. In addition, the survey data suggest that a smaller percentage of inmates smoke in the total-ban facility compared to the partial-ban facility. Taken together, these data suggest that laws banning smoking in correctional facilities can significantly reduce indoor exposure to SHS among inmates, visitors and staff and may reduce prevalence of tobacco use. However, enforcement of smoking bans in combination with cessation services are necessary to realise the full benefit of such policies.

As of November 2007, 24 states have enacted 100% smoke-free correctional facility policies for all indoor areas.9 Studies analysing these bans have found mixed results on efficacy of prison smoking bans as well as inmate specific cessation services.5 10 11 The North Carolina prison study findings are consistent with other studies of indoor air quality after smoking bans were implemented in community settings. In these other community settings the evidence is clear and growing. In Delaware, RSP levels declined similarly in eight hospitality venues after smoking was prohibited there by state law.12 In New York, a similar study observed declining RSP levels in 20 hospitality venues after a smoking ban was put into place.13 However, previous studies of indoor air quality have largely ignored the potential benefits of clean indoor air laws in government controlled buildings such as correctional facilities.

Aside from the rapid reduction in SHS, several studies of the effects of smoking bans suggest that the long-term heath impacts could be substantial as a result of these policies.14 15 A number of studies indicate that respiratory health improved rapidly among workers after smoke-free workplace laws went into effect14 16 Another study by Sargent et al reported a 40% reduction in acute myocardial infarction admissions to a regional hospital during the 6 months that a local smoke-free ordinance was in effect in Helena, Montana.17 18 Similar community studies have been replicated with results ranging from 8% to 39% reductions in Pueblo, Colorado; Bowling Green, Ohio; northern Italy; and most recently New York State.1922

These studies suggest that improvements in air quality and in tobacco-related health effects can occur within months of implementation of a smoking ban. In correctional facilities where tobacco use is very high, non-smoking inmates, staff and visitors cannot avoid exposure to SHS. Such smoking bans protect them from the harmful effects of SHS. The findings in this report are subject to at least five limitations. Firstly, the correctional facilities chosen for this study were not necessarily representative of all prisons in North Carolina or elsewhere. However, they were purposely sampled to provide a range of security levels, inmate genders, sizes and locations. Secondly, SHS is not the only source of indoor particulate matter. Although ambient particle concentrations and cooking smoke are additional sources of indoor particulate levels, SHS is the largest contributor to indoor RSP pollution. Additionally, air-quality monitoring was conducted predominately in dorm areas and lobbies where SHS is the most likely source for concentrations measured. Thirdly, the inmates’ work schedules varied tremendously from area to area and accordingly as did the number of individuals. Therefore, the level of active smoking in any given area at a given time varied at baseline and follow-up. For this reason, PM2.5 concentrations may not accurately represent actual levels. Fourthly, inmates in at least two locations at baseline seemed to think that lower smoking levels could lead to a reversal of the decision to put a smoking ban in place and therefore several inmates may have smoked less during the baseline monitoring. However, this would have spuriously lowered baseline PM2.5 concentrations, which would have biased the results toward the null. Lastly, although smoking bans in the general community have been shown to be efficacious, prison settings are fundamentally different and pose additional challenges, including ethical ones, when enacting evidence-based comprehensive interventions and programmes. This study was not intended to address the ethical or broader challenges facing correctional facilities when implementing indoor air quality laws.


The Center for Disease Control and Prevention’s Best Practices for Comprehensive Tobacco Control Programs focuses on four major goal areas including eliminating non-smoker exposure to SHS.23 The results of this study indicate that a comprehensive statewide ban on smoking in state-controlled buildings such as correctional facilities can significantly reduce SHS exposure indoors. As of November 2007, 17 US states (Arizona, Delaware, Florida, Hawaii, Louisiana, Massachusetts, Minnesota, Montana, Nevada, New Jersey, New York, North Dakota, Ohio, Rhode Island, South Dakota, Utah, Washington) and Puerto Rico currently meet the national health objective for 2010 calling for implementation of statewide smoking bans in worksites.24 These states also account for approximately 54% of the US population. To further reduce the nearly 40 000 deaths among never-smokers caused by SHS each year, similar comprehensive laws are needed in the other 33 states and the District of Columbia. With more than nine million people estimated to be incarcerated worldwide, comprehensive SHS laws and policies would limit indoor SHS exposure and could reduce long-term healthcare costs and adverse health effects of SHS provided that the policies are effectively implemented and enforced and are enacted in combination with evidenced-based cessation services that are tailored to suit the prison setting.

What this paper adds

  • While most air-monitoring studies have historically focused on public venues such as restaurants, to our knowledge this is one of the first air-monitoring studies conducted in state-run prison facilities.

  • Despite pessimism that prison bans are not effective, this study demonstrates that air quality can improve dramatically in a state government controlled prison facility even in a historically tobacco-dependent state where secondhand smoke exposure, smoking rates and smoking attributable healthcare costs are high.

  • Correctional health is an important issue for many countries in terms of healthcare costs, wellbeing of inmates and health of prison employees. Smoking bans can effectively and rapidly reduce secondhand smoke exposure and its adverse health risks to this unique and challenging population.


This research project would not have been possible without the assistance of the following people: Sandra Colt, Donna Dayer, Deborah Dolan, Jennifer Neighbors, Philisa Parker, Marci Paul, Michael Placona of the North Carolina Tobacco Prevention and Control Branch; Mildred Spearman, NC Department of Correction; Andrew Hyland and Mark Travers of Roswell Park Cancer Center; and Kate Uslan of Mecklenburg County Health Department; and the financial support of the Robert Wood Johnson Foundation Substance Abuse Policy Research Program.


View Abstract


  • Funding: Robert Wood Johnson Foundation Substance Abuse Policy Research Program, Rapid Response Grant (No 57808).

  • Competing interests: None.