Perceived Health Impacts, Sources of Information and Individual Actions to Address Air Quality in Two Cities in Nigeria

Abstract

Poor air quality (PAQ) has serious effects on the environment, climate change, and human health. This study investigated the perceived health impacts of PAQ in two cities in Nigeria (Abuja and Enugu), including whether PAQ may have an interaction with COVID-19 infection and intensity. A recent report published in the Lancet has pointed to the complexity of the health care system in Nigeria and a lack of data on disease burden, so the research in this paper took a self-reporting (perceptual) approach to exploring the health impacts of PAQ. The research also sought to explore the main sources of information used by people to inform them about air quality (AQ) and the actions they are likely to take to address PAQ. The results imply that many of the respondents in the two cities perceived their health to be adversely affected by PAQ and that PAQ worsens both the chances of infection and the intensity of COVID-19. Unsurprisingly, older people were found to be more vulnerable to the health impacts of PAQ. Most respondents, especially younger ones, obtained their information on AQ via electronic media (internet, social media) rather than printed media. Respondents considered that the primary action to address PAQ is proper waste management. Paying the government to address PAQ was regarded as the least likely action, although the government was acknowledged as having a key responsibility.

1. Introduction

One of the greatest environmental health risks is caused by PAQ, which adversely affects the health of people, particularly in urban areas [1,2]. The World Health Organization (WHO) has reported that PAQ is associated with 7 million premature deaths annually across the globe [3]. In terms of individual interaction with PAQ, there are potentially a number of different aspects that need to be explored, and these are set out in the relatively simple conceptual model in Figure 1. Firstly, there is how an individual experiences AQ in their immediate environment using perceptual indicators [4,5] and, in particular, how PAQ is perceived to have an impact on their health. Part of this process of framing AQ and its impacts will be influenced by sources of information, such as the media, and discussions with friends, family, and others. There could well be many interactions at play here, of course, as information on AQ may influence how an individual perceives PAQ and perhaps raises awareness as to its potential impacts on health. There is the key question as to what an individual may do about PAQ in terms of trying to mitigate or adapt. There are many actions that could be important here. In Figure 1, they have been grouped into three areas, namely mitigation (reducing causes of PAQ), adaptation (including wearing of facemasks and avoidance of polluted areas), and communication or information (talking with others, including via social media). The ‘action’ choices made by individuals will, in turn, be influenced by information such as that provided in the media and elsewhere. Finally, individual actions after assimilating the information on AQ will improve AQ, reduce exposure to PAQ, and improve health. This shows that such individualized use of AQ data could aid in the creation and execution of unique exposure reduction strategies.

Each of the three main elements of this study (health, information, and action) indicated in Figure 1 has been covered in the literature, especially for people living in the developed world, with an understandable emphasis on the health impacts of PAQ [6,7,8,9,10]. However, in cities of the developing world, much less work has been done on the conceptual model outlined in Figure 1, especially regarding sources of information and the actions of individuals in response to PAQ [11]. At one level, this is unsurprising as health care systems in much of the developing world can comprise a complex web of public, private, third-sector (especially faith-based), and ‘traditional’ institutions, and the same individual may make use of all of these. They may start by visiting a traditional healer, and if that proves unsuccessful, they may go to a clinic or hospital in the public, private, or third sector depending on availability and cost. As a recent report [12] has noted:

“The current health system is sprawling, multifarious, disintegrated, and frequently inaccessible, with very minimal financial risk protection and low financial accessibility of services. Nigerians variously seek care from medical personnel and auxiliaries, community health workers, medicine vendors, marabouts and spiritual healers, traditional birth attendants, and other informal providers. The system relies on a mixture of quasi-tax-funding, fee-for-service, and minimal health insurance coverage.”

Corruption is also endemic in the health care system, and it is plagued by problems with poor data management culture and inadequate research. According to [13,14], the lack of current government data for all indicators inside and outside the health system is evidence that the system as a whole has a low sense of responsibility for its growth and management. As a result, the availability of data on the health impacts of PAQ can be patchy at best, and this needs reform. As [12] has noted:

“Much of Nigeria’s disease burden is uncertain given the near absence of relevant data; for example, in its latest SCORE assessment, WHO estimates that only 10% of deaths in Nigeria are registered. The paucity of data is strongly indicative that decision making is rarely based on appropriate evidence, an enormous challenge that is nonetheless surmountable provided key hurdles are scaled.”

In terms of instrument-based monitoring of AQ, there are also serious issues surrounding the availability of data as such systems are expensive and necessitate technical expertise to operate, maintain, and interpret [4,5]. For that reason, some researchers have proposed the use of perceptual indicators of AQ, and these can map quite well onto what instrument-based data may be available [4]. Hence, this study aimed to explore the three key elements in Figure 1, notably the health effects of PAQ, the sources of information on AQ that people use, and what individuals would do to ensure cleaner air in their vicinities for our case-study cities in Nigeria.

The research reported here is based on a survey and follow-up interviews of respondents in Abuja and Enugu. How individuals perceive AQ and how those perceptions match instrument-based methods of assessment have been addressed in previous publications [4,5]. Here, the focus is on how people perceive the health impacts of PAQ, the sources of information they use and the actions they would take. The paper begins with a brief literature review on these three important aspects of AQ, followed by the methodology and results. The paper ends with a discussion of the findings and their potential impact on policy and management related to AQ.

Figure 1. Conceptual model of the hypothesized relationship between air quality, individual action, and health (adapted and modified from [10]).

Figure 1. Conceptual model of the hypothesized relationship between air quality, individual action, and health (adapted and modified from [10]).

2. Literature Review

2.1. Health Impacts of Poor Air Quality

There is much evidence that PAQ is one of the biggest issues regarding modern urban life. It affects many millions of people every year and even results in millions of fatalities [15]. Mortality from PAQ is typically brought on by exposure to fine particulate matter (PM_2.5) and other air pollutants, which can lead to cancer; cardiovascular diseases including stroke, arrhythmia, heart failure, and hypertension [16,17,18]; and respiratory ailments [1]. PAQ is also linked to dysfunctions of the reproductive and neurological systems [19], skin disorders [20], eye problems [21], and depression [22]. PAQ affects certain demographic characteristics more than others: children, expectant mothers, senior citizens, and people with a history of heart and lung illness are some of those typically most at risk from PAQ. Residents of poor socioeconomic neighborhoods and communities may also be more susceptible to PAQ [23]. Additionally, according to [24], the biological indicators of nervous system function and overall physiological stress may be impacted by metal compounds and the toxicity of tiny particles in dust storms. Numerous risk factors include living close to industrial and transportation air pollution sources, underlying health issues, inadequate nutrition, and stress [23]. Apart from its adverse health effects, PAQ can negatively impact people’s quality of life and can also accelerate climate change and biodiversity loss [25]. PAQ can cause discomfort and anxiety according to [26]. PAQ has been linked to the severity of and susceptibility to COVID-19 [27]. Hazardous air pollutants, specific types of industrial pollution, have been proven to increase the danger, and in some cases the lethality, of COVID-19 [28].

In a study by [29], people were concerned about PAQ mainly because of its adverse health impacts. The study also found that fewer than one-fifth of the respondents voiced their worry about PAQ because they were preoccupied with other topics that were more personally relevant, such as food safety, housing costs, and children’s education. According to [30], regions where respondents thought AQ was getting worse were also areas where they were less concerned about the health effects of pollution, and vice versa for areas where respondents thought air pollution was getting better. The research also noted that places with higher densities of respondents more concerned about the effects of air pollution on their health also had higher PM_2.5 concentrations in the air.

2.2. Sources of Information on Air Quality

As with all environmental issues, having knowledge of AQ helps in terms of mitigation and adaptation [6,31,32]. According to a study conducted on AQ in Muscat, Oman, [33] stated that the primary source of information concerning air pollution was social media. However, the findings identified a considerable difference between men and women, with about three-quarters of women typically obtaining information about AQ through social media, compared with about a quarter who do so through other channels such as newspapers, TV, and radio. Although about 40% of men normally obtain their AQ information from sources other than social media, 60% of men do so. The study also found that 69% of respondents aged 35 to 45 years old reported getting information from social media, which is much higher than the other age groups. There was a considerable disparity in educational attainment in the study; higher educated people (diploma 69%, bachelor’s degree 72%, and postgraduate 59%) and lower educated people (general 55%) obtain AQ information from social media. This shows that higher educated people use social media as a source of information on AQ more than people with lower educational qualifications. A study by [6] reported that there are two distinct sources of information on air quality: public information sources and expert information sources. The reliable public information sources included the US Environmental Protection Agency (EPA), local AQ board websites, email warnings, and apps created by those same organizations for mobile devices that allowed them to access AQ sensor data. A study by [34] on public awareness of PAQ reduction further identified counselling as a way of sensitizing the public to AQ issues.

2.3. Individual Actions in Response to Poor Air Quality

Various studies have explored how individuals and communities respond to PAQ, and many of them revolve around changing energy use and modes of transport. For example, [34] reported responses such as growing plants that reduce PAQ, as well as better maintenance and less usage of petrol and diesel vehicles. According to [35], individuals typically respond to PAQ by saving energy, reducing their car use, avoiding waste burning, sharing rides, using public/mass transportation, or walking to ensure cleaner air. Motivation for behavioral change or implementation of action plans also aids in the reduction of exposure to PAQ [10]. The Minnesota Pollution Control Agency (MPCA) recommended that individuals can act against PAQ by planting and caring for trees, turning off engines when not in use, and becoming clean air champions [36]. However, some people have the opinion that the primary responsibility for addressing PAQ should not be with individuals but with government to set laws and enforce them [37].

3. Materials and Methods

3.1. Research Locations

Nigeria is the most populous nation in Africa. It is located on the coast of West Africa and is bordered by Benin, Niger, Chad, and Cameroon. The capital city of Nigeria, Abuja, is situated in the Federal Capital Territory (FCT), in the geographical center of the country, and Enugu, the capital of Enugu state, is situated in the southeastern part of the country (Figure 2). According to estimates, approximately 217 million people are expected to be living in Nigeria in 2022 [38], with more than half of them residing in urban areas [39].

Abuja is predicted to have a population of around 3.6 million in 2022 [40,41]. At 9°4′ N 7°29′ E and 840 m above sea level, Abuja City has a mean daily temperature of 32.5 °C, with an annual rainfall range of 305–762 mm [42]. Abuja is a relatively new (1970s) planned city with large streets and areas that are designated for residential homes, governmental, and commercial activity. Although there are industries, for instance in the Idu industrial district, the financial service sector, retail, and real estate make up the major elements of Abuja’s economy.

The population of Enugu city is estimated to be 820,000 in 2022 [43]. At 6°27′10″ N 7°30′40″ E and approximately 223 m above sea level, Enugu city has a typical daily temperature of 26.7 °C, with an annual rainfall of about 2000 mm [44,45]. Enugu city is a state capital (Figure 2) and, by way of contrast to Abuja, is an old city, once the headquarters of the Eastern province (1939–1951) and the capital of the Eastern Region following independence from Britain in 1960. It was later the provisional capital of the Republic of Biafra during the Civil War (1967–1970). Enugu is not a ‘well-planned’ metropolis and has no official zoning. As a result, Enugu’s road system has developed in a more piecemeal fashion over many years, and the commercial and Industrial spaces are inter-mixed with residential areas.

3.2. Sample Participants and Demographics

Residents of Abuja (137 respondents) and Enugu (125 respondents) completed structured questionnaires between October 2020 and March 2021, and the sample was stratified to ensure demographic representation of the population profiles in the two cities (

Table 1. Numbers (and percentage) of respondents by demographic characteristics.

Characteristic	Abuja (N = 137)	Enugu (N = 125)
Gender
Male	76 (55%)	63 (50%)
Female	61 (45%)	62 (50%)
Age (cohort)
18–34 years	61 (45%)	60 (45%)
≥35 years	76 (55%)	65 (55%)
Average monthly income ()
0–50,000	57 (42%)	68 (54%)
51,000–100,000	24 (18%)	39 (31%)
≥101,000	56 (40%)	18 (15%)
Highest education qualification
≥Secondary/equivalent	28 (21%)	25 (20%)
Bachelor/diploma or equivalent	59 (43%)	78 (62%)
PhD/master’s or equivalent	50 (36%)	22 (18%)
Main occupation
Unemployed	27 (19%)	21 (17%)
Employed (wage)	93 (69%)	79 (63%)
Self-employed	17 (12%)	25 (20%)
The main modes of transportation used
Walking/cycling	18 (13%)	5 (4%)
Taxi	55 (40%)	46 (37%)
Public mass transit	8 (6%)	52 (41%)
Personal car	56 (41%)	22 (18%)

). In both cities, the stratification produced a roughly 50:50 gender split. Two age categories were used, 18–34 years and 35 years or older, covering 45% and 55% of respondents in each of the two cities, respectively. The division of the population by age into two categories allowed for alignment with Nigeria’s official classification of adults (=>35 years) and children (<35 years) [46]. The minimum age of 18 for respondents was chosen to abide by the research organization’s ethics procedures. Three categories were used for income level: no/low income, mid-income, and higher income. These were selected to provide sufficient numbers of respondents per category for statistical analysis. This use of three income categories in the present study is similar to the lower, middle, and upper-income levels used by [47] in the USA. Of Abuja respondents, 40% were in the higher income category, in contrast with 15% of Enugu respondents in this category. Furthermore, in Abuja, 36% of respondents had higher education (PhD/master’s or equivalent), compared to Enugu, where this was 18% of respondents.

Following the structured questionnaire survey in 2020/2021, follow-up research was conducted in 2022. In this, 20 individuals (10 male and 10 female) for each city were selected from the original questionnaire respondents for in-depth and semi-structured interviews. The sample was equally divided into 5 younger males and 5 younger females (less than 35 years old) and 5 older males and 5 older females (35 years or older) in each of the cities.

3.3. Data Collection and Analysis

Because of the COVID-19 pandemic, the main part of the structured questionnaire survey in 2020/2021 was gathered using the online QualtricsXM^® platform. Hard copies of the questionnaires were used also for those respondents who did not have access to electronic devices. Local field assistants were carefully selected, trained, and supervised to aid the research. During the data collecting period, each of the eight field assistants received an incentive in local currency equivalent to USD 14, and each respondent the equivalent of USD 3 for every properly completed survey form. The field assistants helped ensure proper stratification, the ethical exclusion of people under 18 years of age, and the minimization of the COVID-19 pandemic’s impacts on the demographic categories listed in Table 1. For the in-depth and semi-structured interview phase of the research, conducted between July 2022 and September 2022, 40 respondents were interviewed by the researcher by telephone. Incentives for participation were also provided as USD 10 equivalent to each interviewee and USD 25 equivalent to each field assistant.

Participants in the questionnaire survey were required to provide demographic data, rate the health effects of PAQ, age and gender groups most at risk for PAQ, effects on COVID-19 of PAQ, health effects of PAQ in the previous year, sources of information on AQ, and personal attitudes toward clean air (see Supplementary material S4 for the full questionnaire and

Table 2. Summary of the questionnaire options and Likert scale.

	Likert Scale Scores
Questions	1	2	3	4	5
Health impacts of PAQ	Strongly agree	Agree	Do not know	Disagree	Strongly disagree
Age group most vulnerable impacts of PAQ	Badly impacted	Well impacted	Slightly impacted	No impact	Do not know
Gender most vulnerable impacts of PAQ	Male	Female	Do not know
Does PAQ worsen COVID-19	Yes	No	Do not know
Does PAQ increase the chance of COVID-19 infection	Yes	No	Do not know
Health impact of PAQ in the past year	Very affected	Moderately affected	Slightly affected	Not affected	Do not know
Sources of information on AQ	Extremely important	Very important	Moderately important	Slightly important	Not at all important
Individual actions for clean air	Strongly agree	Agree	Neutral	Disagree	Strongly disagree

for the summary). On Likert scales of 1 to 5 and 1 to 3, survey participants were asked to rate their overall perception of the questions just for the city in which they resided (Table 2). The questionnaire was piloted to ensure its feasibility for the survey. The ranking of the answers was based on these overall perceptual values. SPSS^® v28.0 software was utilized for the statistical analysis. The Kruskal–Wallis and Hochberg post hoc tests were used to examine differences in scores between demographic groups and questions to find groupings that were statistically homogeneous at p < 0.05 [48,49]. The rank order based on the y-axis data is indicated by numbers next to locations on the x-axis. The 95% confidence interval for the means is shown by the error bars. The categories determined by the Hochberg post hoc test at p < 0.05 (5%) are represented by the numbered horizontal bars (A = Abuja, E = Enugu).

In the second semi-structured interview phase of the research, respondents were asked follow-up questions about the topics covered in the original survey. The interviews were recorded and transcribed, and the texts were subjected to content analysis with coding. Results are presented here primarily in the form of quotes, and respondents have been coded using the following: Aym, Aom, Ayf, and Aof for Abuja respondents and Eym, Eom, Eyf, and Eof for Enugu respondents; A = Abuja, E = Enugu, y = younger (18–34 years) o = older (35 years and above), m = male, and f = female. Some of the quotes by the respondents are relevant to more than one of the issues investigated in this study.

4. Results

4.1. Health Impacts of Poor Air Quality

Figure 3a,b give the mean scores and 95% confidence intervals of the adverse health effects of PAQ perceived by respondents in Abuja and Enugu. The health effects are ordered so that those with the greatest perceived health effects of PAQ are on the left-hand side of the x-axes of Figure 3, while those having the least effect on health are on the right-hand side. The Hochberg post hoc test was employed to divide the health impacts into statistically homogenous groups (p < 0.05), and four groups were identified in both cities, with codes A1–A4 for Abuja and E1–E4 for Enugu. While there are some minor differences, the respondents in Abuja and Enugu have very similar perceptions of the health impacts of PAQ, with respiratory diseases (mean scores: 1.69 in Abuja and 1.70 in Enugu) perceived as the main impact of PAQ, and depression (mean scores: 2.79 in Abuja and 2.86 in Enugu) perceived to be least impacted.

Table 3. Kruskal–Wallis test results for the perceived health impacts of PAQ for groups based on some demographic characteristics in Abuja and Enugu.

Health Impacts of PAQ	Cities	Gender	Age	Income	Education	Occupation	Transportation
Respiratory diseases	Abuja	2.728	6.524	4.265	3.858	4.205	12.568 **
	Enugu	5.456	8.251	1.528	11.159 *	10.676 *	2.553
Eye/skin problem	Abuja	2.053	3.660	2.092	4.088	1.360	3.463
	Enugu	4.435	9.414 *	2.637	5.985	2.335	2.854
Heart diseases	Abuja	2.551	7.748	11.406 *	7.198	5.012	6.822
	Enugu	1.643	4.754	2.675	11.362 *	3.599	1.345
Low immune system	Abuja	11.659 *	3.505	9.720 *	7.478	1.294	6.510
	Enugu	1.822	3.095	1.058	7.452	2.072	2.773
Cancer	Abuja	2.299	2.679	3.167	0.841	0.744	0.548
	Enugu	4.502	6.772	4.288	9.798 *	1.850	2.236
Depression	Abuja	1.878	2.333	3.442	2.044	2.987	2.698
	Enugu	7.806	3.999	2.175	4.920	5.781	3.030

Note: Values shown are the Kruskal–Wallis statistic; * p ≤ 0.05, ** p ≤ 0.01. Highlighted cells indicate the presence of significant differences (p < 0.05) within the demographic group.

presents the results of Kruskal–Wallis tests designed to explore differences between demographic groups in terms of their scoring of the health impacts of PAQ. The table is arranged from the greatest perceived impact on health at the top of the table to the lowest perceived impact on health at the bottom. Statistically significant differences (at p < 0.05) between demographic group responses were identified for only a few of the health issues; out of 72 tests in Table 3, only 9 of them were statistically significant at p < 0.05. Income and education were the only demographic characteristics with more than two significant statistical differences. For example, the results from Enugu suggest that higher educated people are more aware than the less educated and mid-educated people that PAQ leads to respiratory diseases, heart diseases, and cancer. These answers from respondents regarding the importance of education also emerged in the follow-up interviews of phase 2 as the following quotations illustrate:

“I want to believe that in this case, it is education. Most of these things are taught in schools. Education plays a major role here. The younger ones are more educated than the older ones in Enugu.”

—Eof1.

“The understanding of the meaning of air quality. The awareness of the air quality of the cities. The more educated and higher income earners are more aware or exposed to what air quality implies and can know why air quality is poor more than the less educated and lower income earners.”

—Aom7.

4.2. Age Groups Perceived to Be most Affected by Poor Air Quality

Figure 4a,b illustrate that older people of 75 years and above (mean scores: 1.91 in Abuja and 1.58 in Enugu) were perceived by the respondents in Abuja and Enugu to be the most affected age group. People who are 18 to 24 years were identified to be least affected in Abuja, while in Enugu the least affected age group was perceived to be those who are 25 to 34 years.

Table 4. Kruskal–Wallis test results for differences between age groups most affected by PAQ in Abuja and Enugu.

Age Group Most Affected by PAQ	Cities	Gender	Age	Income	Education	Occupation	Transportation
Less than 18 years	Abuja	1.418	2.364	5.809	2.921	3.996	7.540
	Enugu	2.153	7.362	2.828	8.676	8.917	4.913
18–24	Abuja	0.871	3.537	12.083 *	9.264	0.861	5.234
	Enugu	2.997	5.787	3.537	1.670	5.933	3.371
25–34	Abuja	0.703	3.061	6.403	9.087	1.379	9.140
	Enugu	0.516	9.120	3.222	5.028	3.546	2.806
35–44	Abuja	1.771	6.026	9.401	5.627	4.030	5.337
	Enugu	2.813	12.397 *	2.793	4.981	2.576	2.766
45–54	Abuja	3.167	6.052	11.413 *	2.884	1.437	3.978
	Enugu	2.276	3.741	1.371	4.582	4.522	9.145
55–64	Abuja	1.928	10.528 *	12.546 **	5.217	1.235	2.574
	Enugu	2.480	2.382	2.060	4.051	2.458	4.736
65–74	Abuja	6.593	7.012	8.087	2.556	10.855 *	12.172 *
	Enugu	1.694	3.863	2.170	5.122	5.825	5.061
75 years and above	Abuja	1.961	8.049	3.336	11.820 *	2.972	6.049
	Enugu	3.452	0.856	2.129	2.127	2.811	1.733

Note: Values shown are the Kruskal–Wallis statistic; * p ≤ 0.05, ** p ≤ 0.01. Highlighted cells indicate the presence of significant differences (p < 0.05) within the demographic groups.

shows how the demographic groups perceived the adverse impacts of PAQ on age groups. Few, just 8 out of 96 tests, were statistically significant at p < 0.05, suggesting there was broad agreement between these demographic groups as to the age groups that were most and least susceptible to PAQ. The few statistically significant differences were mostly in the income (3 significant differences) and age (2 significant differences) demographic characteristics and were only observed in Abuja. Income and education are related, of course, as people with higher education tend to have higher income levels.

The lower income earners in Abuja (see S3) assigned higher scores to the age group impacts by PAQ. It will be of interest that the lower income earners are mainly people of younger age groups, and this might have influenced their opinion on the age groups that are most affected by PAQ in Abuja. Interestingly, this may not be attributed to higher awareness and education, which, contrastingly, were identified in Section 4.1 as the main influencing factors for PAQ perception. These higher scores of the age group impacts by PAQ may not be connected to their economic status, their residential area, and occupation as the lower income earners tend to work or live in areas of poorer AQ. According to the viewpoint of one of the respondents:

“The higher income earners and higher educated people are more aware of poor air quality due to their education and where they live.”

—Aof9.

4.3. Gender Perceived to Be most Affected by Poor Air Quality

In both cities, most participants said they did not know which gender, if any, would be most affected by PAQ (Figure 5). Of those who did make a choice, females (36% and 28% in Abuja and Enugu, respectively) were identified to be the most affected by PAQ in both cities, compared to males (19% in Abuja and 18% in Enugu). As shown in

Table 5. Kruskal–Wallis test results for differences between some demographic groups perceived to be most affected by PAQ in Abuja and Enugu.

Cities	Gender	Age	Income	Education	Occupation	Transportation
Abuja	2.818	1.382	4.405	1.153	0.359	2.336
Enugu	8.840 **	1.572	0.297	0.331	1.653	1.359

Note: Values shown are the Kruskal–Wallis statistic; ** p ≤ 0.01. Highlighted cells indicate the presence of significant differences within the demographic group.

, there is only one statistically significant difference out of 12 tests on how demographic groups perceive different impacts of PAQ on gender. The only statistically significant difference is among the gender groups in Enugu (S1), suggesting that men perceived females to be most affected by PAQ, while women did not think this was the case.

4.4. Interaction between Poor Air Quality and COVID-19

The majority (>60%) of respondents in both cities believe that PAQ aggravates the severity of symptoms experienced by individuals infected with COVID-19 (Figure 6). Figure 7 illustrates that a majority of respondents in both cities also believe that PAQ can lead to a greater chance of COVID-19 infection. In Enugu, 61% of the respondents agreed that PAQ can increase the chance of being infected by the virus, 16% did not agree, and 23% did not know if PAQ leads to a greater chance of catching coronavirus. In Abuja, some 47% believed that PAQ could increase the chances of getting COVID-19, while 26% did not agree, and 27% said they did not know.

4.5. Perceived Health Effects of Poor Air Quality on Individuals in the Past Year

Many of the respondents believed that PAQ had affected their health negatively in the year (2019–2020) prior to the survey (Figure 8). Results suggest that 74% and 75% in Abuja and Enugu, respectively, believed that PAQ had negatively impacted their health over that time, while the health of 8% in Abuja and 13% in Enugu were said to have been seriously affected by PAQ, and only 19% and 11%, respectively, believed that PAQ had not affected their health.

Table 6. Kruskal–Wallis test results for perceived differences between some demographic groups in terms of health effects of PAQ during 2019/2020 (the year prior to the survey) in Abuja and Enugu.

Cities	Gender	Age	Income	Education	Occupation	Transportation
Abuja	0.835	5.057	3.423	5.216	6.017	6.591
Enugu	17.296 **	12.077 *	2.637	9.198 *	1.293	5.527

Note: Values shown are the Kruskal–Wallis statistic; * p ≤ 0.05, ** p ≤ 0.01. Highlighted cells indicate the presence of significant differences (p < 0.05) within the demographic group.

illustrates that there are three demographic groups with statistically significant differences in terms of the perceived impact of PAQ on health in the year prior to the survey. These differences only occurred in Enugu. The results suggest that women, older people, and higher educated people in the city were the groups perceived to be most affected by PAQ over the previous year. This may in part be a matter of awareness linked to education and experience. For instance, the statements by the following respondents, as well as that of Aom7 in Section 4.1, are linked to awareness of and experience in AQ in the area:

“Those with higher income and higher education are more aware of what contributes to poor air quality than the less educated and lower income earners who do not know much about air quality.”

—Ayf8

“Many older people in Enugu who have had the experience of how the city environment used to be or has changed with many activities can say the air quality is poorer than before while the younger who do not have much of experience would say the environment or air quality is better.”

—Aom4.

“The older people are more affected or vulnerable and they may rate air quality poorer in the Enugu city, and the younger ones may not be as concerned about their air quality like the older people.”

—Ayf5.

“The older people must have lived in the area when there were less population and activities that contribute to poor air quality. That is to say that the older ones are more experienced with the locations and that is why they rated the areas poorer than the younger ones.”

—Eof6.

4.6. Sources of Information on Air Quality

Figure 9 illustrates the sources from which the residents of Abuja and Enugu obtained their information on AQ. For most residents of both cities, this was primarily via the internet (mean scores: 2.04 in Abuja and 2.06 in Enugu). Indeed, the responses suggest that printed media—leaflets, newspapers, magazines, and so on—were the least important resources, while electronic media were the most important. In terms of differences between the demographic groups,

Table 7. Kruskal–Wallis test results for differences between groups based on some demographic characteristics and sources of information on AQ in Abuja and Enugu.

Source of Information	Cities	Gender	Age	Income	Education	Occupation	Transportation
Internet	Abuja	3.632	2.623	4.823	0.527	2.568	3.251
	Enugu	3.527	12.022 *	3.276	6.258	5.036	2.649
Television	Abuja	5.539	6.706	6.019	7.860	0.257	4.265
	Enugu	3.296	5.632	0.496	9.278*	8.598	10.550 *
Social media	Abuja	11.122 *	8.738	3.368	2.022	2.683	7.484
	Enugu	2.753	18.540 **	4.123	5.762	12.053 **	6.745
Radio	Abuja	1.245	10.736 *	2.288	0.994	1.748	1.153
	Enugu	2.076	1.912	4.820	10.640 *	8.633	4.027
Friends	Abuja	5.670	9.319 *	7.111	6.136	5.705	8.400
	Enugu	4.974	13.406 **	0.853	2.985	4.919	4.758
Newspapers	Abuja	0.578	3.248	5.371	2.574	3.925	8.907
	Enugu	2.617	4.874	3.245	1.847	6.326	5.421
Posters	Abuja	11.695 *	1.839	7.199	2.099	4.873	5.668
	Enugu	5.067	10.389 *	4.172	2.143	6.355	3.053
Leaflets	Abuja	3.282	3.790	3.852	2.764	5.218	4.057
	Enugu	1.796	5.337	1.875	2.967	5.204	1.002
Magazines	Abuja	1.029	1.582	10.811 *	6.174	3.307	5.629
	Enugu	1.795	3.004	1.712	4.633	5.819	5.261

Note: Values shown are the Kruskal–Wallis statistic; * p ≤ 0.05, ** p ≤ 0.01. Highlighted cells indicate the presence of significant differences (p < 0.05) within the demographic group.

shows that only 13 out of the 108 tests were significantly different at p < 0.05, and half of these (6) were linked to age groups in Enugu.

Table 8. Mean scores, standard deviations (SD), and Kruskal–Wallis (KW) test results for age and sources of information on AQ for Abuja and Enugu.

		Abuja			Enugu
Source of Information	Age Groups	Mean	SD	KW	Mean	SD	KW
Internet	18–34	1.92	0.988	2.623 ns	1.79	1.051	12.022 *
	35 and over	2.13	1.215		2.31	1.139
Television	18–34	2.03	0.912	6.706 ns	2.30	1.243	5.632 ns
	35 and over	2.07	1.124		2.55	1.284
Social media	18–34	1.97	1.032	8.738 ns	1.89	0.968	18.540 **
	35 and over	2.24	1.165		2.58	1.066
Radio	18–34	2.28	0.878	10.736 *	2.51	1.260	1.912 ns
	35 and over	2.22	1.207		2.50	1.182
Friends	18–34	2.49	1.043	9.319*	2.44	1.162	13.406 **
	35 and over	2.47	1.270		3.03	1.098
Newspapers	18–34	2.75	1.150	3.248 ns	2.75	1.299	4.874 ns
	35 and over	2.71	1.335		3.22	1.240
Posters	18–34	3.00	1.270	1.839 ns	3.02	1.335	10.389 *
	35 and over	2.96	0.988		3.22	1.133
Leaflets	18–34	3.15	1.152	3.790 ns	3.07	1.263	5.337 ns
	35 and over	2.99	1.160		3.25	1.069
Magazines	18–34	3.15	1.263	1.582 ns	3.25	1.274	3.004 ns
	35 and over	3.04	1.270		3.53	1.221

Note: Figures in the table are the Kruskal–Wallis statistics; * p ≤ 0.05, ** p ≤ 0.01; ns = not significant. Highlighted cells indicate the presence of significant differences (p < 0.05) within the age demographic groups by sources of information on AQ.

provides a breakdown of the sources of information across the two age groups in Abuja and Enugu. The results suggest that in Enugu, younger people identified the internet, social media, friends, and posters as their main sources of information on AQ more than did older people. These findings were confirmed in the interviews as illustrated by the following sample of quotations:

“At the moment, the younger ones are having more access to information on air quality due to the new technology coming up than the older ones. They have more access to the internet and are more aware of the control measures than the older ones.”

—Eym12.

“Miseducation has made the older generation 35+ and above not decode the need for clean air and the ills from air pollution. Younger people even without witnessing clean air or a good ecosystem have seen it on the internet which they access more than the older generation and are more educated to understand that it could have been better than what it is in the present”

—Eym3.

Interestingly, the relative homogeneity in the perception by the age groups in Abuja was mainly linked to city status and awareness. For instance, a respondent stated the following:

“I believe that there is no difference in Abuja because Abuja is a developed city. Everybody tends to believe in the same thing as most of the activities by old and young are almost the same way. There are not many ghettos in Abuja but there are in Enugu.”

—Ayf3.

4.7. Individual Actions for Clean Air

Figure 10 illustrates the actions that respondents said they were most likely to take towards the improvement of AQ in Abuja and Enugu. In both cities, the most likely action would be to properly dispose of waste (mean scores: 1.54 in Abuja and 1.52 in Enugu), while the least likely action would be to pay money to the government (mean scores: 3.52 in Abuja and 3.65 in Enugu) to improve AQ. In Abuja, the second most important action was the planting of trees, but in Enugu, it was the avoidance of activities that are considered to cause a decline in AQ. It is interesting to note that saving energy and the use of mass transport systems scored relatively low in both cities and were clearly not priorities for the respondents. There were few differences between the demographic groups in their responses. The low importance of paying money to the government to address PAQ is an interesting point since the interview respondents noted that keeping the air clean is the main responsibility of the government:

“It is the government that should be in the prime position to control poor air quality.”

—Eyf4.

“It is the primary responsibility of government to implement air pollution control, but it is a collective responsibility of both citizens and government to control poor air quality with organized public knowledge, sensitization & orientation”.

—Eym3

“In this case, I think the government should take the primary responsibility. The government should set rules for managing poor air quality and should be held responsible”.

—Eof5.

It would seem people are less willing to pay the government to address PAQ, yet this would appear to be at odds with their point that government is the key actor in addressing the issue.

5. Discussion

The research reported here addresses key elements of public perceptions and responses to PAQ in two case-study cities in Nigeria as set out in the conceptual model of Figure 1. In terms of the perceived impacts of PAQ on health, studies such as [50] in Abuja and [51] in Enugu have noted some of the general health impacts of air pollutants, but the research reported here is the first to identify residents’ perceptions on how their health and wellbeing may be adversely affected by PAQ in these Nigerian cities, along with how PAQ may interact with COVID-19. The findings from the two cities in terms of perceived health impacts were very similar, and there were few differences between demographic groups. Indeed, the results of this study are broadly in line with the findings of many others in that respiratory diseases were perceived to be the greatest impact of PAQ [1,17,52,53,54,55,56,57,58,59,60,61,62,63,64]. Depression was seen by respondents as being the least negative health effect of PAQ. In terms of the demographic groups, the few observed statistical differences could be attributed to awareness or knowledge of the topic as higher-educated people are more positive in their identification of some of the health effects of PAQ [S2]. The study identified that the most vulnerable age group to the negative impacts of PAQ was perceived to be the elderly (=>75 years), and this result supports previous studies [22,65,66] that found older adults to be the age group that suffers most from the health impacts of PAQ, largely due to their weaker immune system and lower level of physical activity or exercise. Both women and men were perceived to be affected by PAQ, although a minority of respondents felt that women may be more negatively affected. This latter finding is consistent with aspects of the work of [67], which reported that women are more vulnerable to PAQ because of longer exposure. The results also clearly indicate that PAQ was perceived to aggravate the adverse effect of COVID-19, and most respondents believed that PAQ enhances the chance of someone being infected with the coronavirus. These outcomes support the findings of [68,69,70,71], which reported that PAQ worsens the effects of COVID-19.

In terms of information sources, the finding that the internet was the most important source of information on AQ in both cities is in agreement with other research [72,73]. Social media was also identified to be one of the most important sources of information on AQ, and this has also been noted by others [74,75,76]. The least used means of AQ information dissemination according to the present research are devices such as posters, leaflets, newspapers, and magazines. This suggests a clear advantage for electronic sources of information on AQ relative to the more traditional ‘printed’ media. Inadequate awareness of PAQ has been noted as an important issue by various authors (e.g., [37,77]), and the results from this study suggest that agencies need to focus their efforts on the use of electronic media rather than printed forms.

For actions that people are most likely to take, it is intriguing to note the emphasis on waste disposal by respondents in both cities. There are other studies that have noted the importance of waste disposal for PAQ [78,79,80,81], but it is interesting to see this action ranked above all other options, including energy saving and choice of transport, which often feature prominently in the literature [35,82,83,84,85]. Indeed, it is encouraging to note how respondents placed emphasis on individual action to help clean their air instead of paying the government to do it. Yet there does appear to be a seeming contradiction, since in the follow-up interviews it was clear that respondents perceived the government to be the key actor responsible for addressing PAQ. The apparent difference in response may well be due to important aspects such as trust. While respondents regard the government as the key actor in addressing PAQ, they are reluctant to pay the government to address PAQ as they may well feel the money would not achieve the desired outcomes, perhaps because of inefficiencies and corruption. Hence, the findings suggest that while the responsibility of government should not and cannot be ignored, there should nonetheless be a focus on the promotion of, and support for, individual actions against PAQ in both cities.

There are significant issues with the availability of health data in Nigeria, and this needs to be addressed with urgency [12,13,14]. The same is also true for instrument-measured data related to AQ in Nigeria [4,5]. Until such data become available, more studies that take a self-reporting approach to the health impacts of PAQ, such as the one reported here, are required, but the literature on such perceptual approaches to the investigation of health impacts of PAQ, means of information on AQ, and individual attitudes to mitigating PAQ are limited for cities in the developing world. The present research pioneers this form of study in Abuja and Enugu in Nigeria and has helped to address that important gap. The approach taken is likely to have relevance in other areas of the world where less research on the issue has been conducted. This study has also illustrated the degree of awareness and the perceptions of people on the negative effects of PAQ and provides policymakers, civil societies, and individuals with evidence to help guide policies and actions to improve PAQ, health, and other environmental issues.

6. Conclusions

This research was conducted using a representative sample of respondents from the case-study cities. The following conclusions were reached:

• The beliefs of residents on the health impacts of PAQ are similar in two different developing-world cities (Abuja and Enugu) in Nigeria.
• The perceptions of people in the study areas are mainly because of awareness and knowledge of AQ.
• The majority of the residents believe that they are affected by PAQ and were affected in the year prior to the research (74% in Abuja and 75% in Enugu).
• Older people are identified to be the group of people most vulnerable to the deleterious effects of PAQ (mean scores: 1.91 in Abuja and 1.58 in Enugu).
• PAQ was perceived to make COVID-19 worse and to increase the chances of COVID-19 infection and aggravate the effects of the pandemic.
• The Internet is the major source of information about AQ in Abuja and Enugu (mean scores: 2.04 in Abuja and 2.06 in Enugu).
• Action on proper waste disposal was identified by residents as the most highly ranked option to reduce PAQ (mean scores: 1.54 in Abuja and 1.52 in Enugu).
• Residents considered PAQ an issue for the government but were unwilling to pay the government to manage PAQ.