Introduction
Well-being is a multifaceted construct and while there is no consensus on a single definition, the Centers for Disease Control and Prevention (CDC) describes well-being as “the presence of positive emotions and moods, the absence of negative emotions, satisfaction with life, fulfillment, and positive functioning [Reference Chutiyami, Cheong and Salihu1].” The interest in studying well-being within health research has drastically increased over the last 20 years. Using the PubMed database, there were 1,361 results using the term well-being in 2003, in 2022 there were 22,536 results for the term well-being. While use of the term has increased, we have not seen the same attention applied to defining the term comparably across fields of study. Colloquially, well-being is often defined or discussed as a synonym for wellness, health, happiness, and satisfaction. Within the academic community, we define well-being as a multifaceted construct with definitions that vary by domain. For example, the definition of emotional well-being will differ from the definition of physical well-being or economic well-being. Although aspects of well-being seem universal, how it is depicted in the literature has substantial variation in definition and even greater variation in how it is measured.
Specifically, within the field of occupational health and well-being, we have also seen an increase in the interest in measuring and improving workers’ well-being. In 2011, the National Institute of Occupational Safety and Health within the CDC expanded the traditional delivery of occupational safety and health by integrating well-being [2]. Total Worker Health® was introduced as a strategy that combines health protection with health promotion to prevent worker injury and advance well-being [Reference Chari, Chang and Sauter3]. The recent coronavirus disease 2019 pandemic has brought even greater attention to the importance of worker well-being. Much concern has been specifically expressed about the mental health and well-being of healthcare professionals during and at the height of the pandemic [Reference Chutiyami, Cheong and Salihu1]. However, psychological distress from the pandemic on the overall workforce has led to greater turnover intention [Reference Poon, Lin, Griffiths, Yong, Seah and Liaw4], resignation [Reference Jiskrova5], and ultimately, labor shortages. So much so that in 2022, the U.S. Surgeon General released a new framework for mental health and well-being in the workplace, stating that it is “a critical priority for public health [6].” Protection from harm, connection and community, work-life harmony, mattering at work, and opportunities for growth were the five essentials that were highlighted to guide leaders in developing an organizational culture that supports worker mental health and well-being [6]. Therefore, the purpose of this paper is to identify conceptual definitions and operational assessments of well-being within the field of occupational health.
Methods
Studies were identified by searching PubMed September 2022 and April 2023 using the search terms “well-being,” “occupational OR workplace,” and “scale.” The inclusion criteria were (1) scholarly journal articles, (2) published in English, (3) measured well-being, and (4) empirical papers. The search was not limited by date of publication. From this search, 4394 articles were identified. After reviewing the abstracts, 3733 articles were removed for not having a well-being measure, leaving 661 articles for full review. Four reviewers conducted the screening using pre-established inclusion criteria. In the first screening, reviewers independently screened the abstracts for inclusion criteria. If one reviewer indicated an article as relevant at the initial screening phase, the article proceeded to the second screening to ensure inclusivity. In the second screening, reviewers independently screened the full text of articles to ensure the articles met the inclusion criteria. Following a full review of the articles, 273 papers were removed, leaving 388 articles included in the data extraction for the present study.
Data Extraction
The following information was extracted from the 388 articles that met the inclusion criteria: (a) Well-being assessment citation, (b) name of the well-being assessment, (c) items and rating, (d) reliability and validity, (e) samples/populations from occupational health and well-being studies, (f) assessment limitations noted in the occupational health and well-being studies, and (g) other well-being assessments used in combination with this assessment. This information was condensed by assessment so that the same assessment was only listed once. This information is displayed in Table 1.
Results
There was a total of 158 well-being assessments that resulted from our scoping review of the occupational health and well-being literature. The full table is available as an online supplemental file. A condensed version of the table with the most relevant well-being assessments (N = 21) can be found in Table 1. Of the information extracted, three sections are highlighted in this section (1) reliability and validity of the well-being assessments, (2) the samples/populations represented within the occupational health and well-being articles included in this review, and (3) limitations of the assessments noted in the included studies.
Reliability and Validity
The most common psychometric information reported was internal consistency or reliability represented by Cronbach’s alpha (α). There were a few single-item measures and items rated as yes/no, for which reliability cannot be tested. The Cronbach’s alpha coefficient is meant to provide a measure of the internal consistency. The coefficient ranges between 0 and 1, with a value of 0.70 or higher indicating good consistency and reliability [Reference Tavakol and Dennick7]. While reliability should be determined before validity, reliability alone does not tell us if the measure is also valid. We found that few assessments reported the validity and for those that did, the type of validity varied (i.e., predictive, discriminant, convergent, etc.). Therefore, we have little information about whether the assessments that claim to measure well-being are valid measures of the construct.
Populations
There were a variety of samples and populations represented across assessments. While we limited our study to only articles published in English, there were a range of countries represented across studies. We also had a variety of employment types reported across studies, with the majority being from healthcare or healthcare-related fields.
Limitations
We extracted the author-reported limitations of their study using the well-being assessment. The listed limitations were specific to the potential impact of the assessment of well-being. The most frequent limitations mentioned were the lack of generalizability of the well-being findings to other populations and small sample sizes. These are both subjective to the author’s perspective, but we believe these are worth considering when choosing a well-being assessment appropriate for each study.
Discussion
The overall lack of attention to the measurement and assessment of well-being and use of inconsistent types of measures of well-being in published articles is concerning. Most studies resulting from our search did not properly report how they assessed well-being. A deficiency in the use of a standard definition may, in part, explain the heterogeneity of well-being measurements that were reported. Utilization of a standardized definition and shared conceptual framework may help researchers develop strong measurements that accurately depict and report well-being.
Populations Represented in Occupational Health and Well-Being Literature
A unique feature of our scoping review was the extraction of information related to the samples and populations that have been included when measuring occupation/workplace well-being. We purposely allowed for a wide variation in populations to gather information regarding international assessments of well-being, but we were limited by only being able to review studies published in (or translated to) English. At first glance, we did not notice a difference in how well-being was assessed between cultures, but future studies may be able to use the data we extracted and presented in Table 1 to perform a more formal analysis to assess potential differences in well-being assessments between cultures.
Recommendations for Assessing Well-Being
Based on the extensive review of over 300 articles, we have developed three recommendations for researchers who want to improve their well-being assessment. First, we were surprised at the number of articles that had to be excluded following a full review because they did not actually measure well-being despite discussing well-being in the introduction sections and having concluding remarks in their discussion sections. We recommend that authors do not mention well-being unless they have measured it and if they use a composite of measures, we recommend they explain how the composite operationalizes well-being. Second, be as precise as possible in your conceptual definition of well-being. We saw multiple articles that used a broad conceptual definition of well-being, but then a specific and narrow operational definition or assessment tool. We recommended researchers introduce a specific definition of well-being (e.g., economic, emotional, physical, spiritual) in their introduction section that will help the readers understand which domain of well-being is being assessed. We recommend using subscales or focused scales when measuring specific domains of well-being, such as emotional well-being. We do not recommend concluding emotional well-being based on an overall well-being assessment. We suggest that the term well-being only be used when multiple constructs are used together to assess an overall composition of well-being, beyond what can be captured through a single aspect of well-being. It is clear across conceptual definitions that well-being is a multifaceted construct that cannot be captured through a single dimension. When using a single construct to assess well-being, we suggest defining the individual construct rather than using the single construct to define well-being. For example, if a researcher is measuring quality of life, happiness, and health, those combined measures could be used to infer well-being, or they may be using a multidimensional well-being scale. But, if the study is only measuring quality of life, then the researcher should only infer quality of life, not well-being. Across fields, it is imperative to procure validated instruments that accurately measure well-being and reflect participants’ data accurately.
Limitations and Future Research
The results of the present study should be interpreted with the following limitations in mind. Due to the overwhelming large scope of research that could be included with the simple term “well-being,” we are not able to present the full body of research in a single scoping review, therefore we decided to focus specifically on occupational well-being. The present study was limited to empirical studies indexed in PubMed and published in the English language only.
We would also like to acknowledge the potential issue of false positive and false negative when searching for articles that measure well-being. There may have been false positive results by including articles that do not directly measure well-being, but conclude well-being based on proxy measures of mental and physical health. These articles are falsely included because the authors use the term “well-being” and therefore the article was found during our searches. Additionally, there may have been false negative results by missing articles that did not use the term “well-being” to describe their results, but based on their measure, we would have defined their construct as measure of well-being. Because the authors did not use the term well-being, their article was not a result of our searches.
There were many articles that discussed well-being in the introduction and discussion but did not measure well-being in the methods and results. For the purposes of this review, these papers were excluded as they did not provide adequate explanation of the measurement of the well-being construct. Future research may be interested in looking at this issue more specifically and what it means for the field to conclude well-being or make implications for well-being without measuring the construct directly.
Selecting the appropriate assessment of well-being for each study is a challenge and there is currently no standard process for selecting the best assessment tool. This is a promising future avenue of work for researchers interested in creating a flow chart to assist researchers in finding an assessment that fits their study aims and methods. There are currently online repositories hosted by groups such as The University of Connecticut (UConn) M3EWB (Mechanisms Underlying Mind-Body Interventions and Measurement of Emotional Well-Being) Network that allows researchers to find assessments for specific types of well-being. For example, researchers can search these repositories for an emotional well-being assessment for children. These repositories, if maintained, can be an excellent tool for managing the most reliable and valid assessments in the field. We believe the table available as a supplement file and the condensed table presented in this paper are also useful tools for researchers to use to identify a well-being assessment tool that fits the needs of their study. These tables may also be used for future analyses to search for patterns and gaps in current measurement. For example, someone may use these tables to see if there are common limitations across assessments or the most common combination of well-being assessments or look for missing populations and use existing assessments within those populations.
There is a need to clearly define and differentiate the term well-being from other constructs to create measures that adequately capture the importance of the term and its antecedents. Assessing if and how well-being differs by cultures and sample characteristics, such as age, education, race and ethnicity, and clinical profile (e.g., disease/disorder, problem severity, comorbidity), could provide valuable insights to improve translational science.
Conclusion
The current review highlighted the inconsistency of research examining the measurement of well-being. Additional research is needed to develop rigorous measurements of well-being that can be used across study populations and adequately capture the multiple dimensions of well-being. There is a need to provide consistent definitions and precise language when inferring well-being from results.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/cts.2023.648.
Funding statement
This project was supported by The University of Rochester Center for Leading Innovation and Collaboration as the coordinating center for the Clinical and Translational Science Awards Program, funded by the National Center for Advancing Translational Sciences at the National Institutes of Health; U24TR002260 and KL2TR002542. Individual authors would also like to acknowledge their time supported by other mechanisms: R34DA057150 (LMD), P30AI073961(LMD). R25AT010664 (TGB).
Competing interests
The authors have no conflicts of interest to declare.