Knowledge–Behavior Gap in Tap Water Consumption in Puerto Rico: Implications for Water Utilities

Public mistrust is sometimes treated as synonymous with distrust or misplaced trust (Breakwell 2020) and refers to the uncertainty about whether trust should be offered. Public mistrust, in this paper, refers to a lack of confidence and belief in the safety and quality of the water provided by public water utilities such as PRASA. Without trust or belief in a system, every effort made by water utilities to provide potable water that end users will consume would prove futile. Public mistrust affects social behavior by creating a sense of risk; however, there may not be conclusive evidence of the effects of these risks (Breakwell 2020; Calman 2002). Literature shows examples of promising water interventions that proved ineffective because there was public mistrust of project engineers from water utilities (Borland 2014; Harvey and Reed 2006). Some studies have found a direct correlation between the mistrust of tap water providers or bottled water companies and the public consumption rates from these sources (Doria 2006). A separate study also found that university students who trusted their local water utility to deliver safe drinking water were likelier to drink from tap water sources. In contrast, those who mistrusted their government and university were more likely to drink bottled water (Grupper et al. 2021; Saylor et al. 2011).

The sources of public mistrust of tap water include individual and household indicators of socioeconomic status such as education level, household income, and racial or ethnic minority status (Pierce et al. 2019; Pierce and Gonzalez 2017). However, regardless of individual and household indicators of socioeconomic status, people are more likely to mistrust their tap water when it is unpalatable than when it is unsafe (Pierce and Lai 2019; Spackman and Burlingame 2018). The palatability of tap water is measured by its aesthetic characteristics, such as color, odor, and taste. A slight change in the color, odor, or taste of drinking water may raise suspicion (Young et al. 1996). In the United States, water quality is governed by primary and secondary regulations (CDC 2022). Primary regulations limit contaminants that may impact human health (USEPA 2015b), while secondary regulations guide parameters associated with aesthetic qualities such as taste, color, and odor (USEPA 2015a). Traditionally, it is expected that adequately treated water should not only meet primary standards for health safety but also align with secondary standards, thereby presenting little to no color, taste, or unpleasant odor (Kearns et al. 2015). However, even when health-based standards are met, aesthetic issues can still arise because secondary standards may not be regulated or enforced. These aesthetic qualities can significantly influence public perception and trust in water utilities, even when the water is safe to consume from a health standpoint. The palatability ratings drive alternative drinking consumption, particularly bottled water consumption preferences over tap water (Huerta-Saenz et al. 2012). These palatability perceptions are important in distinguishing between tap and household filtered water drinkers, even when respondents have similar perceptions about unfiltered tap water safety, contamination, and health risks (Triplett et al. 2019). While literature assesses some sources of public mistrust for drinking water, there is a gap in why Puerto Ricans may express mistrust in tap water, especially after the post-Hurricane Maria recovery process. Puerto Rico’s vulnerability to climatic events such as Hurricane Maria, which damaged its water infrastructure managed by PRASA, coupled with the socioeconomic vulnerabilities of residents who rely heavily on tap water from PRASA, makes it essential to study why there is public mistrust of tap water in Puerto Rico.

Puerto Rico tends to perform worse in providing safe drinking water than the mainland United States due to a combination of a lack of investment in infrastructure, maintenance, testing, and the effects of natural disasters (Hunter et al. 2010; Karim et al. 2020; Keenum et al. 2021). Hurricanes and tropical storms have made it difficult to ensure water quality meets USEPA standards on the island (Ferré et al. 2019; Kaufman 2019; Keenum et al. 2021; Sánchez-Colón et al. 2022). In 2021, of the 78 municipalities in Puerto Rico, 17 were considered severe violators of the Safe Drinking Water Act by the USEPA. On the other hand, 400 out of the 456 drinking water utilities in Puerto Rico have had some violations within the last 3 years (Mueller and Gasteyer 2021). It is not uncommon to find tap water with high levels of fecal coliform contamination (Holman et al. 2014) and heavy metals (Apeti et al. 2012; Ortiz-Colón et al. 2016) in conjunction with unappealing palatability characteristics such as foul odor and color (Gonzalez 2002; Jain et al. 2014). While the water quality in some municipalities may be better than others, the available studies demonstrate that water quality in many municipalities is not up to USEPA standards. Researchers have linked poor water quality to severe health implications for residents. Water-related diseases such as diarrhea have increased because of the poor water quality on the island (Ferré et al. 2019; Ghosh et al. 2021; Hunter et al. 2010). While there is extensive knowledge among scientists about the poor piped water conditions supplied by PRASA, it is unclear to what extent the population of Puerto Rico is aware of the health implications of the poor water quality. Moreover, although a survey has not been done to measure the population’s awareness, it is possible to infer that people may be aware but choose to consume tap water regardless of the potential health risks, because more than half the population on the island has used tap water as a drinking water source as of 2017 (Lin et al. 2020; Michaud and Kates 2017).

In the literature on mistrust in the water sector, the predominant theory has often focused on the concept of trust and its erosion (Grupper et al. 2021; Pierce et al. 2019; Pierce and Gonzalez 2017). Trust has been considered a vital element in public perceptions and behaviors related to tap water consumption (Grupper et al. 2021). Previous studies have explored the factors contributing to the breakdown of trust, such as water contamination incidents, challenges with water infrastructure, and perceived inadequacies in water management practices (Grupper et al. 2021; Pierce et al. 2019; Pierce and Gonzalez 2017).

Trust, conceptually speaking, can be understood as the belief or confidence that a person or entity is dependable and will act in a manner that aligns with expectations (Blomqvist 1997; Gedge and Abell 2020; Pierce and Gonzalez 2017). It involves a willingness to be vulnerable and place faith in something or someone. However, trust may not always lead to the expected behaviors, as demonstrated by the example of people trusting tap water but not consuming it. While trust theories have been valuable in shedding light on tap water mistrust, they have not fully addressed the knowledge–behavior gap. This gap refers to the disparity between what individuals know about water safety and their actual behaviors regarding tap water consumption. The disconnect between awareness and behavior implies that trust theories have primarily focused on exploring the psychological and social factors influencing trust in tap water (Grupper et al. 2021; Pierce and Gonzalez 2017). While these factors are essential in shaping people’s attitudes, they might not be the sole determinants of whether individuals actually act on that trust (de França Doria 2010). By highlighting the knowledge–behavior gap, our study seeks to bridge this theoretical limitation.

The knowledge–behavior gap theory, on the other hand, is a concept in the field of public health that posits that individuals may have knowledge about healthy behaviors, such as the importance of regular exercise or eating a balanced diet, but they may not engage in those behaviors (Ajzen 1991; Michie et al. 2011; Sligo and Jameson 2000). The literature on this theory has been growing in recent years, with many studies focusing on identifying the factors contributing to the gap between knowledge and behavior. Some of the key factors that have been identified include social and environmental factors, such as access to healthy food options or safe places to exercise, and individual factors, such as lack of self-efficacy (confidence in one’s ability to carry out a behavior) or time constraints (Jones et al. 2009; Sligo and Jameson 2000). Additionally, the knowledge–gap behavior may be fueled by health beliefs, such as perceived susceptibility to a particular health problem; health literacy, including the understanding of health information and ability to use it in decision-making; and psychological factors, such as lack of motivation or habit formation (Jones et al. 2009; Sligo and Jameson 2000).

In this study, our primary objective was to investigate the factors contributing to the mistrust of tap water among residents in Puerto Rico, as well as to examine their behaviors in response to or despite this mistrust. The context of Puerto Rico after Hurricane Maria presents a compelling backdrop for exploring the knowledge–behavior gap phenomenon. The island has experienced previous incidents of water contamination and challenges with water infrastructure (Ghosh et al. 2021; Michaud and Kates 2017; Preston et al. 2020), which may have eroded trust in the safety and quality of tap water. Considering the existing literature on climate hazards and water management, there remains a critical gap in comprehending public mistrust of tap water, particularly in the aftermath of a disaster. Our study seeks to address this gap by specifically focusing on the post-Hurricane Maria context in Puerto Rico, where the devastation severely impacted the island’s water infrastructure (Marcos 2022). We employ the theory of the knowledge–behavior gap within the water management sector context.

The theory of the knowledge–behavior gap, often utilized in health behavior research, explores the disconnect between individuals’ knowledge and their actual behaviors. By integrating the knowledge–behavior gap theory with trust/mistrust perspectives in the context of tap water mistrust in Puerto Rico, our research aims to provide a more comprehensive understanding of the factors influencing attitudes and behaviors. This approach recognizes the importance of trust in shaping perceptions and actions, while also considering the role of knowledge and behavior alignment in the postdisaster setting. By exploring the reasons behind the persistence of tap water mistrust in Puerto Rican residents, despite efforts to restore water quality after hurricane Maria, we aim to shed light on the unique experiences, perceptions, and challenges faced by the population in relation to their attitudes and behaviors toward tap water. This knowledge can inform targeted interventions and strategies that address both trust dynamics and the knowledge–behavior gap, ultimately promoting healthier water consumption practices in Puerto Rico.

Furthermore, our research contributes to the broader field of interdisciplinary research. By examining tap water mistrust from various perspectives, including public health, sociology, psychology, environmental science, and engineering, we embrace an interdisciplinary approach. This collaboration allows us to gain a comprehensive understanding of the multifaceted factors influencing tap water mistrust and the knowledge–behavior gap. By integrating insights from different disciplines, we develop more holistic and effective strategies to address tap water mistrust and promote behavior change. This contribution highlights the importance of interdisciplinary research in tackling complex issues related to water mistrust and emphasizes the need for collaboration across diverse fields to achieve meaningful and impactful outcomes.

Our study contributes to the broader literature on mistrust in the water sector by providing a comprehensive understanding of the intricate relationship between perceived knowledge and behavior, particularly within the context of tap water consumption. We aim to address the limitations of existing trust theories by shedding light on these interconnected factors and their implications. First, our research focuses on understanding the knowledge–behavior gap related to tap water consumption behaviors. By investigating this gap, we offer insights into the reasons for mistrust in tap water and how this mistrust impacts actual behaviors. We explore whether trust in tap water sources consistently translates into tangible actions, such as consumption, or whether there are other factors that hinder this translation. Second, our research focuses on residents in Puerto Rico, who represent a vulnerable population heavily reliant on public utilities, especially in the aftermath of disasters like Hurricane Maria. By studying this specific context, we gain valuable insights into the unique challenges faced by this population and their responses to tap water mistrust. This knowledge is crucial for developing targeted interventions and strategies to address the issues surrounding mistrust in tap water, particularly in vulnerable communities.

We applied mixed research methods to explore the tap water consumption behaviors and perceptions among residents of three municipalities in Puerto Rico. By combining surveys and interviews, we aimed to capture a comprehensive understanding of the residents’ experiences and views. In sum, our study contributes to a more nuanced understanding of public mistrust in tap water, addressing a critical gap in the water sector literature. The insights gained from our study have the potential to inform policy and practice, improving water security for communities grappling with climate-related hazards and aging infrastructure.

Our three study areas are Loíza, Comerío, and Aguas Buenas. Loíza is a densely populated, predominantly Black-Hispanic municipality about 39 km (24 mi) east of the capital city, San Juan. Comerío and Aguas Buenas are predominantly White-Hispanic municipalities about 44 km (27 mi) and 32.7 km (20 mi), respectively, south of San Juan. The three municipalities were chosen based on the considerable damage to the water infrastructure from Hurricane Maria. After Hurricane Maria, all three communities spent close to two months without access to safe drinking water. Furthermore, these communities still experience an unsteady supply of safe drinking water to their homes due to faulty piping infrastructure and frequent power outages (Laskow 2018; Marcos 2022).

We conducted the surveys and interviews simultaneously. To qualify for an interview and survey, participants had to have: (1) been 18 years old and above; and (2) experienced Hurricane Maria with flooding on their streets or property with a subsequent change in their drinking water quality. Both requirements were to ensure that residents had experienced how tap water quality was before Hurricane Maria. We went through a comprehensive process of explaining the informed consent form in the language the research participants were most comfortable with. Research participants were also provided a $25 gift card. One of four research assistants trained in research methods administered all surveys and interviews in Spanish and in person. The Institutional Review Board (IRB) of Iowa State University approved our project methods.

We adopted a narrative analysis approach to analyze the data collected from the interview transcripts (Bamberg 2012). This approach, known for its capability to locate narratives in context and extract meaningful insights from them, was apt considering the unstructured and open-ended nature of our interviews. Our unit of analysis was each individual participant’s account regarding their water consumption behavior and their trust or mistrust in tap water. Following an inductive analysis approach, we allowed themes to emerge naturally from the data instead of relying on predetermined categories (Yilmaz 2013). We utilized a qualitative coding framework to categorize and group these emergent themes. The initial codes were developed based on patterns observed in the data and then refined iteratively throughout the analysis process. The central themes from the interviews were grouped based on whether or not people drink tap water. These two groups were further divided into those who trust tap water and those who mistrust tap water.

In our study, three individuals were involved in the coding and analysis process. The initial coding and translation of the interviews from Spanish to English were conducted by two of the authors. One of the authors, a professor in urban planning, is a native Puerto Rican. Additionally, a professor in civil engineering and their PhD student in civil engineering who collected the data in the field also participated in this phase. Regarding the coding process itself, we employed a mixed approach to coding, with PhD student coder utilizing NVivo, a qualitative data analysis software, while the other two professor coders opted for manual coding. The coding across the three coders allowed for a comparison between the two methods and added an additional layer of rigor to the coding process. The first coder who used NVivo leveraged the software’s features to organize, code, and analyze the interview data. NVivo facilitated a systematic and efficient approach to coding, enabling the coder to assign codes, link-related segments, and visualize patterns and themes within the data set. On the other hand, the other coders chose to manually code the transcripts without using NVivo. This approach involved reading through the transcripts, identifying key concepts, and manually assigning codes to macro- and microcodes. The manual coding process required a closer and more intimate engagement with the data, as the coder familiarized themselves with the content and carefully selected appropriate codes based on their interpretation of the interview responses.

To assess the agreement between the coders, a systematic approach was adopted. In the initial meetings, the team members reviewed a subset of the coded data to compare their coding decisions and identify any discrepancies. Through this collaborative process, consensus was reached on the coding framework, and any discrepancies were discussed and resolved (O'Connor and Joffe 2020). The team continued to refine their coding approach until a high level of agreement was achieved (Hallgren 2012). Ultimately, the decision was made to consider all codes aligned, and substantial agreement was reached on both the assigned codes and the corresponding quotes.

The coding and analysis of the data continued until no new codes and themes emerged, and theoretical saturation was achieved, following the approach outlined by Glaser and Strauss (2017). The concept of theoretical saturation refers to an analytical technique that indicates that the existing interviews provide sufficient information for theory development. Throughout this process, instances of negative cases or interviews that did not align with the overall coding framework were carefully considered by the authors, leading to reflection and subsequent modifications to the coding scheme. Saturation was deemed to have been reached when the identified codes and themes demonstrated repeated occurrences during the analysis of new interviews.

Table 1 summarizes the distribution of the demographic information across the interviews. It also shows each community’s percentages of tap water trust and consumption patterns. Table 2 shows the distribution of the qualitative themes from the interviews across all three communities.

Group A represents the broad category of residents in Puerto Rico who mistrust the quality of tap water provided by PRASA. Out of the 154 interviewed, 99 (64%) residents expressed mistrust toward their tap water. Out of that number, 85 (86%) did not drink tap water entirely. This observation is expected as mistrust in water sources is associated with nonconsumption (Juran and Lahiri-Dutt 2017; Kooy and Walter 2019). Of residents who expressed mistrust in tap water and did not drink it, Loiza had 33 (39%), while Comerio observed 27 (32%), and Aguas Buenas, 25 (29%) (Table 1). This difference in mistrust levels could be attributed to the fact that Loiza, located near the coast of Puerto Rico, has faced challenges with water quality in the past. The municipality has had issues with contamination and infrastructure problems, particularly after hurricane Maria. These events can erode trust in the local water supply and make residents more cautious about consuming tap water. Comerio and Aguas Buenas, both mountainous areas, on the other hand, might have had fewer reported incidents or a better track record in terms of water quality. The coding framework shown in Table 2 for Group A revealed that for residents who did not drink the tap water as a result of their mistrust, their primary reasons for mistrust were the aesthetics of tap water characterized by bad taste, color, and smell and the perceived health risks associated with drinking tap water. On the other hand, for residents who drink tap water despite expressing their mistrust, their primary reasons for mistrust were the utility’s poor treatment processes and the substandard piping systems of the utility.

Group B represents the broad category of residents in Puerto Rico who trust in the quality of tap water provided by PRASA. A total of 55 (35.7%) residents out of the 154 interviewed trusted their tap water. A total of 40 of the 55 residents who trusted their tap water consumed it. The coding framework shown in Table 2 for Group B revealed that for residents who trust and drink their tap water, their primary reasons for trusting it are good experiences from decades of usage and the utility’s excellent treatment of the tap water. Similarly, residents who did not drink tap water despite reporting that they trusted it had the same reasons as their counterparts who drank tap water. These residents also trusted their tap water because they had good experiences from decades of usage and believed that the utility treated the tap water excellently.

The phenomenon of misaligned knowledge–behavior relationships among a group of residents among Group A2 residents presents a perplexing dilemma. Despite evidence of reasons for mistrust in the local water supply, such as poor piping systems, these residents continue to consume tap water. The context of Puerto Rico must be taken into consideration when examining this phenomenon. It is possible that financial constraints and lack of accessibility to alternative drinking water sources may play a role in this knowledge–behavior gap (Cortés 2018; Delilah Roque et al. 2020; Yu et al. 2015). It has been observed that residents are less likely to seek out alternative sources of drinking water due to mobility and logistical difficulties (Apt 2013; Banks et al. 2019; Wrisdale et al. 2017). These extenuating circumstances highlight the importance of ensuring that tap water quality meets acceptable federal standards and that the information is readily accessible to residents. Furthermore, this knowledge–behavior gap may have negative psychological implications, as residents are aware of the potential health hazards associated with the consumption of unsafe water yet continue to consume it due to a lack of feasible alternatives. Further research is recommended in order to understand and address this enigmatic behavior and ensure that residents are not forced to compromise their health due to financial constraints or other factors.

On the other hand, in an interesting reversal of the typical knowledge–behavior gap, some residents in Puerto Rico demonstrate trust in the safety of their tap water while refraining from consuming it. Despite having confidence in its quality, individuals may opt for alternative sources of drinking water due to the accessibility of other alternative water sources, such as bottled water, or external pressure from family or friends over the concerns of tap water-quality issues. This phenomenon highlights the complex relationship between knowledge, perception, and behavior and underscores the importance of addressing not only water-quality issues but also broader systemic challenges to ensure residents' access to safe and reliable drinking water.

Group A1 consisted of residents who believed that their tap water was unsafe for consumption and, as a result, did not use it. This sentiment is shared by a significant portion of the Puerto Rican population, as evidenced by ongoing concerns about the injustices experienced in the aftermath of Hurricane Maria. The aesthetic characteristics of the tap water, specifically appearance, smell, and taste, were identified as primary concerns among residents of Loíza, Comerío, and Aguas Buenas. It can be inferred that if these characteristics had been improved, a larger portion of the population might have continued to consume tap water. This observation is significant as it suggests that negative public perception and mistrust of tap water may be largely influenced by first-hand experiences with the water’s appearance, smell, and taste, a correlation that has been previously established in the literature (de França Doria 2010; Pierce and Gonzalez 2017). Additionally, residents’ concerns about the potential health hazards associated with tap water consumption were found to be validated by the literature, with studies documenting a high incidence of waterborne diarrheal diseases in Puerto Rico and other Caribbean islands (Hunter et al. 2010). However, it is important to note that further research is needed to determine the accuracy of these negative perceptions and to understand the extent of water contamination in Puerto Rico.

On the other hand, Group B1 represents the situation where residents trust in tap water quality and use it as their primary drinking water source. This behavior is significant because, despite the challenges and issues faced, the tap water in most communities in Puerto Rico generally meets the established drinking water criteria as set by regulatory standards (Mueller and Gasteyer 2021). However, it is important to note that there have been instances where drinking water criteria have been violated, particularly in specific locations or during certain periods of time (NRDC 2017; Michaud and Kates 2017). Also, the way in which the public is informed of problems with water-quality when/if they occur plays a role in the types of perceptions formed about tap water (Kim et al. 2023; Lucier et al. 2020). This is particularly important because of the increasing number of vulnerable people in Puerto Rico, and therefore, the need for reliable information for tap water consumption decision-making is greater.

PRASA’s monopoly status plays a significant role in water utility management in Puerto Rico. The lack of competition could result in a lack of motivation for PRASA to improve service quality and efficiency. Possible alternatives to this monopoly include decentralized water management or community-based water provision, which could offer more flexibility and adaptability, especially in rural areas. These decentralized systems have been successful across Añasco, Mayaguez, and Rincón, where water-sharing networks are used to overcome PRASA’s inefficiencies (Roque et al. 2021). Such systems could potentially lead to more sustainable water practices and a smaller knowledge–behavior gap.

The issue of the knowledge–behavior gap, as identified in this study, is a nuanced problem that needs a tailored approach to resolve. This gap is essentially a divergence between consumers’ understanding of tap water quality provisioned by water utilities and their behavior in relation to it. The findings of this study reveal that while consumers may exhibit trust toward tap water, their consumption patterns often tell a different story. This discrepancy indicates that personal experiences with water utilities are a significant factor in shaping consumer behavior. Therefore, it becomes essential to address this knowledge–behavior gap in utility management.

To address the knowledge–behavior gap, the first recommendation would be to augment public education on the potential risks associated with contaminated tap water. However, utility providers may be limited by budgetary and institutional constraints. Therefore, the emphasis should be on making the best use of available resources to maximize educational outreach. Second, the study suggests that community-wide initiatives should be fostered to tackle tap water-quality issues. However, it is crucial to be mindful of the practical implications of such efforts, because utilities often operate under tight constraints financially. Third, there is a need for policies that strengthen the protection of tap water quality. While formulating and implementing such policies, it is essential to consider the realities of limited revenue and governmental support. Policies should be designed in such a way that they are effective despite these constraints. Last, it is recommended that an improved, robust system of routine monitoring and reporting of water consumption behaviors and quality be instituted. This would hold PRASA and other responsible entities accountable. The complexities of such an undertaking should not be underestimated, but a system of checks and balances is crucial for effective utility management.

While it is important to acknowledge that water utilities often function in the context of complex settings, financial constraints, and environments, these strategies, while designed to tackle the identified knowledge–behavior gap, are also meant to take into consideration the realities and constraints of water utility management. By following these, utilities can not only bridge the gap but also ensure improved consumer satisfaction and efficient use of resources.