Threats to the Internal Validity of Experimental and Quasi-Experimental Research in Healthcare.

The article defines, describes, and discusses the seven threats to the internal validity of experiments discussed by Donald T. Campbell in his classic 1957 article: history, maturation, testing, instrument decay, statistical regression, selection, and mortality. These concepts are said to be threats to the internal validity of experiments because they pose alternate explanations for the apparent causal relationship between the independent variable and dependent variable of an experiment if they are not adequately controlled. A series of simple diagrams illustrate three pre-experimental designs and three true experimental designs discussed by Campbell in 1957 and several quasi-experimental designs described in his book written with Julian C. Stanley in 1966. The current article explains why each design controls for or fails to control for these seven threats to internal validity.

The t-test: An Influential Inferential Tool in Chaplaincy and Other Healthcare Research.

The t-test developed by William S. Gosset (also known as Student's t-test and the two-sample t-test) is commonly used to compare one sample mean on a measure with another sample mean on the same measure. The outcome of the t-test is used to draw inferences about how different the samples are from each other. It is probably one of the most frequently relied upon statistics in inferential research. It is easy to use: a researcher can calculate the statistic with three simple tools: paper, pen, and a calculator. A computer program can quickly calculate the t-test for large samples. The ease of use can result in the misuse of the t-test. This article discusses the development of the original t-test, basic principles of the t-test, two additional types of t-tests (the one-sample t-test and the paired t-test), and recommendations about what to consider when using the t-test to draw inferences in research.

Hospital Characteristics Affecting HealthCare Chaplaincy and the Provision of Chaplaincy Care in the United States: 2004 vs. 2016.

This study replicates, expands and analyzes a 2004 survey examining six hospital characteristics influencing three measures of chaplain employment in large, small, for-profit and nonprofit hospitals. The relationship between hospital characteristics and hiring Board Certified Chaplains was minor and inconsistent across time. The results indicate that religiously affiliated hospitals employed more full-time chaplains and that chaplain full-time equivalents were inversely related to hospital size in both surveys. The current survey suggests that urban and religiously affiliated hospitals were more likely to hire chaplains. The sampling method proved problematic, precluding meaningful conclusions but the study focus and questions remain important for future investigation based on this pilot effort.

International Study of Chaplains' Attitudes About Research.

An online survey was conducted by twelve professional chaplain organizations to assess chaplains' attitudes about and involvement in research. A total of 2,092 chaplains from 23 countries responded to the survey. Over 80% thought research was definitely important and nearly 70% thought chaplains should definitely be research literate. Just over 40% said they regularly read research articles and almost 60% said they occasionally did. The respondents rated their own research literacy as 6.5 on a 0-10 scale. Significant positive inter-correlations were found among all four measures: importance of (a) research and (b) research literacy; (c) frequency of reading articles; and (d) research literacy rating. Approximately 35% were never involved, 37% had been involved, 17% were currently involved, and 11% expected to be involved in research. The last three groups were significantly more likely to think research and research literacy were important and to read research articles than chaplains who were never involved in research. Given chaplains' interest in research, actions should be undertaken to facilitate further research engagement.

Studying Associations in Health Care Research.

This article discusses some of the types of relationships observed in healthcare research and depicts them in graphic form. The article begins by explaining two basic associations observed in chemistry and physics (Boyles' Law and Charles' Law), and illustrates how these associations are similar to curvilinear and linear associations, respectively, found in healthcare. Graphs of curvilinear associations include morbidity curves and survival and mortality curves. Several examples of linear relationships are given and methods of testing linear relationships with interval and ratio data are introduced (i.e., correlation and ordinary least-squares regression). In addition, 2 × 2 contingency tables for testing the association between categorical (or nominal) data are described. Finally, Sir Austin Bradford Hill's eight criteria for assessing causality from research on associations between variables are presented and explained. Three appendices provide interested readers with opportunities to practice interpreting selected curvilinear and linear relationships.

Measures of variability in chaplaincy, health care, and related research.

This article discusses statistical measures of variability in relation to measures of central tendency and levels of measurement. Three measures of variability used in healthcare research (the range, the interquartile range, and the standard deviation) are described and compared, including their uses and limitations. The article describes how each of the three measures is calculated, and it provides a step-by-step example of calculating the sums of squares, variance, and standard deviation. Graphs of frequency and percentage distributions are used to show how the interquartile range and the standard deviation represent the variability observed within distributions. The article discusses the properties of the normal curve regarding the distribution of scores around the mean in relation to the standard deviation, and illustrates differences in the shapes of normal curves with the same mean but different standard deviations.

Measures of central tendency in chaplaincy, health care, and related research.

The three measures of central tendency are discussed in this article: the mode, the median, and the mean. These measures of central tendency describe data in different and important ways, in relation to the level of measurement (nominal, ordinal, interval, or ratio) used to obtain the data. The results of published research studies, thought experiments, and graphs of frequency and percentage distributions of data are used as examples to demonstrate and explain the similarities and differences among these summary measures of data. The examples include the application of nominal, ordinal, interval, and ratios scales to measure pain, anxiety, chaplaincy services, religious behaviors, and treatment-related preferences, and their respective measures of central tendency. Examples of unimodal and bimodal distributions, and differences in the relative locations of the median and mean in symmetrical and skewed distributions are also presented and discussed.

Independent, dependent, and other variables in healthcare and chaplaincy research.

This article begins by defining the term variable and the terms independent variable and dependent variable, providing examples of each. It then proceeds to describe and discuss synonyms for the terms independent variable and dependent variable, including treatment, intervention, predictor, and risk factor, and synonyms for dependent variable, such as response variables and outcomes. The article explains that the terms extraneous, nuisance, and confounding variables refer to any variable that can interfere with the ability to establish relationships between independent variables and dependent variables, and it describes ways to control for such confounds. It further explains that even though intervening, mediating, and moderating variables explicitly alter the relationship between independent variables and dependent variables, they help to explain the causal relationship between them. In addition, the article links terminology about variables with the concept of levels of measurement in research.

Fundamentals of measurement in health care research.

This article discusses levels of measurement and their application to research and practice in health care. The concept of levels of measurement was codified in a seminal article by S. S. Stevens in 1946 that defined four levels of measurement: nominal scales, which label and classify cases (objects and individuals) and assign them to categories; ordinal scales, which rank cases on some attribute; interval scales, which have equal intervals for measuring attributes; and ratio scales, which have equal intervals and a natural zero point. The rules that apply to each level of measurement are presented and the mathematical operations that can be performed on them are explained. The similarities and differences among the four types of scales are discussed and examples of their use in health care and other contexts are described.

Operational definitions in research on religion and health.

This article summarizes the historical development of operational definitions and discusses their application to research on religion and health, and their importance for research, in general. The diversity of religious concepts that have been operationalized is described, as well as the development of multi-dimensional self-report measures of religion specifically designed for use in health research. The operational definitions of a variety of health concepts are also described, including the development of multi-dimensional self-report measures of health. Some of the most consistently observed salutary relationships between religion and health are mentioned. The rising interest in spirituality in health research is discussed, along with problems with the current operational definitions of spirituality in healthcare research. The levels of measurement used in various, operationally defined religious and healthcare concepts are highlighted.

Research designs and making causal inferences from health care studies.

This article summarizes the major types of research designs used in healthcare research, including experimental, quasi-experimental, and observational studies. Observational studies are divided into survey studies (descriptive and correlational studies), case-studies and analytic studies, the last of which are commonly used in epidemiology: case-control, retrospective cohort, and prospective cohort studies. Similarities and differences among the research designs are described and the relative strength of evidence they provide is discussed. Emphasis is placed on five criteria for drawing causal inferences that are derived from the writings of the philosopher John Stuart Mill, especially his methods or canons. The application of the criteria to experimentation is explained. Particular attention is given to the degree to which different designs meet the five criteria for making causal inferences. Examples of specific studies that have used various designs in chaplaincy research are provided.

Beliefs about God and mental health among American adults.

This study examines the association between beliefs about God and psychiatric symptoms in the context of Evolutionary Threat Assessment System Theory, using data from the 2010 Baylor Religion Survey of US Adults (N = 1,426). Three beliefs about God were tested separately in ordinary least squares regression models to predict five classes of psychiatric symptoms: general anxiety, social anxiety, paranoia, obsession, and compulsion. Belief in a punitive God was positively associated with four psychiatric symptoms, while belief in a benevolent God was negatively associated with four psychiatric symptoms, controlling for demographic characteristics, religiousness, and strength of belief in God. Belief in a deistic God and one's overall belief in God were not significantly related to any psychiatric symptoms.

Patients' expectations of healthcare chaplaincy: a cross-sectional study in the German part of Switzerland.

Identifying patients' expectations of and need for healthcare chaplaincy is important in terms of appropriate intervention. Therefore, a sample of 612 patients from 32 general hospitals and psychiatric clinics in the German part of Switzerland was surveyed about their expectations of chaplaincy service. A principal component factor analysis of participants' ratings found that the survey items fell into three distinct categories. These were the need for (1) emotional support, (2) help to cope with illness/disease, and (3) religious/spiritual assistance. Among the expectations, the need for emotional support was rated most important, followed by help to cope and, lastly, religious/spiritual assistance. Gender, religious denomination, general religiosity, and subjective health status significantly influenced these expectations. The results showed that fulfilling patients' expectations increases their overall satisfaction with, and the importance they accord to the chaplain's visit, as well as their confidence in the chaplain.

A national study of chaplaincy services and end-of-life outcomes.

BACKGROUND: Medicine has long acknowledged the role of chaplains in healthcare, but there is little research on the relationship between chaplaincy care and health outcomes. The present study examines the association between chaplaincy services and end-of-life care service choices. METHODS: HealthCare Chaplaincy purchased the AHA survey database from the American Hospital Association. The Dartmouth Atlas of Health Care database was provided to HealthCare Chaplaincy by The Dartmouth Institute for Health Policy & Clinical Practice, with the permission of Dartmouth Atlas Co-Principal Investigator Elliot S. Fisher, M.D., M.P.H. The Dartmouth Atlas of Health Care is available interactively on-line at http://www.dartmouthatlas.org/. Patient data are aggregated at the hospital level in the Dartmouth Atlas of Health Care. IRB approval was not sought for the project because the data are available to the public through one means or another, and neither database contains data about individual patients, i.e. all the variables are measures of hospital characteristics. We combined and analyzed data from the American Hospital Association's Annual Survey and outcome data from The Dartmouth Atlas of Health Care in a cross-sectional study of 3,585 hospitals. Two outcomes were examined: the percent of patients who (1) died in the hospital, and (2) were enrolled in hospice. Ordinary least squares regression was used to measure the association between the provision of chaplaincy services and each of the outcomes, controlling for six factors associated with hospital death rates. RESULTS AND DISCUSSION: The analyses found significantly lower rates of hospital deaths (ß = .04, p < .05) and higher rates of hospice enrollment (ß = .06, p < .001) for patients cared for in hospitals that provided chaplaincy services compared to hospitals that did not. CONCLUSIONS: The findings suggest that chaplaincy services may play a role in increasing hospice enrollment. This may be attributable to chaplains' assistance to patients and families in making decisions about care at the end-of-life, perhaps by aligning their values and wishes with actual treatment plans. Additional research is warranted.

Belief in life-after-death, beliefs about the world, and psychiatric symptoms.

Data from the 2010 Baylor Religion Survey were analyzed by structural equation modeling (SEM) to test five hypotheses: (1) that religious commitment is positively related to belief in life-after-death; that belief in life-after-death is (2) positively related to belief in an equitable world, and (3) negatively related to belief in a cynical world; (4) that belief in a cynical world has a pernicious association with psychiatric symptoms; and (5) that belief in an equitable world has a salubrious association with psychiatric symptoms. As hypothesized, religious commitment was positively related to belief in life-after-death (ß = .74). In turn, belief in life-after-death was negatively associated with belief in a cynical world (ß = -.16) and positively associated with belief in an equitable world (ß = .36), as hypothesized. SEM further confirmed that belief in a cynical world had a significant pernicious association with all five classes of psychiatric symptoms (ß's = .11 to .30). Belief in an equitable world had a weaker and less consistent salubrious association with psychiatric symptoms. The results are discussed in the context of ETAS theory.

Religion, forgiveness, hostility and health: a structural equation analysis.

Religious participation has been shown to increase certain factors thought to be protective of health, including social support and positive health habits. The current study considers whether religious participation may likewise have a positive influence on health by increasing forgiveness and diminishing hostility. A structural equation analysis of data collected from a national survey of 1,629 participants supported the hypothesized model that (a) religiosity is related to greater forgiveness, (b) greater forgiveness, in turn, is related to reduced hostility and finally, (c) reduced hostility is related to better subjective health.