Patient Satisfaction and Management

¶ … diagnoses, pain is a common complaint among inpatients. In the U.S. alone, approximately 100 million patients experience chronic pain (Alaloul et al., 2015). Pain negatively affects numerous aspects of an individual's life, such as sleep, quality of life, and physical functioning. Pain is also associated with negative psychological outcomes like depression, extended hospitalization, and a huge economic burden. In the U.S., for instance, pain imposes an estimated cost of $635 billion on patients and the healthcare system as a whole (Alaloul et al., 2015). Ineffective management of pain can have a negative impact on patient satisfaction, underscoring the need for more effective interventions.

Effective pain management is particularly important in postpartum care, where the experience of pain is common (Eshkevari, Trout & Damore, 2013). However, the management of pain in postpartum care remains quite ineffective, with up to 20% of postpartum patients reporting dissatisfaction with pain management (Niemi-Murola et al., 2007; Espenshade & Hreniuk, 2017). The dissatisfaction stems from, among other factors, poor midwifery care, little or individualized care, poor communication between providers and patients, and lack of involvement in decision making (Mohammed, 2016). For mothers, such experiences may lead to negative childbirth experiences. Accordingly, it is crucial for nurses in postpartum care to use more effective pain management interventions so as to increase maternal satisfaction with pain management.

Generally, structured nurse rounding is one of the ways through which patient satisfaction in nursing practice can be improved (Mitchell et al., 2014; Brosey & March, 2015; Alaloul et al., 2015). This intervention essentially involves monitoring the patient on a regular basis. For instance, the patient can be assessed every one hour during the day and every two hours during the night. The assessment focuses on a number of patient aspects such as pain, temperature, safety hazards, positioning, and need for toileting (Brosey & March, 2015). Patient satisfaction can also be increased by improving provider-patient communication (Singh et al., 2011; Tan et al., 2013; Alaloul et al., 2015). Can these techniques improve patient satisfaction in postpartum care? Based on this premise, the following PICOT question is formulated: In postpartum, women undergoing either vaginal or caesarean delivery (P), does hourly rounding and use of whiteboard (I), compared to no rounding and no use of whiteboard (C), increase patient satisfaction with pain management (O) in a 3-month trial period (T)? The following section provides a critique of five quantitative research articles relating to the PICOT question.

Critique

Brosey, L., & March, K. (2015). Effectiveness of structured hourly nurse rounding on patient satisfaction and clinical outcomes. Journal of Nursing Care Quality, 30(2), 153-159.

Research Purpose

The purpose of the study was to monitor the impact of structured hourly nurse rounding on patient falls, hospital acquired pressure ulcers (HAPU), and patient satisfaction. The nursing intervention was implemented at a 24-bed medical-surgical unit within a large community hospital.

Research Design

The study took the form of a descriptive study. The descriptive design is one of the major forms of quantitative research designs. Essentially, a descriptive study aims to describe the characteristics of a given phenomenon, meaning that the researcher does not formulate a hypothesis at the beginning of the study (Creswell, 2014). Also, a descriptive study usually does not involve randomization, blinding, or a control group. The chosen design ideally fits the nature of the study, which was aimed at examining the effectiveness of structured hourly nurse rounding on patient satisfaction. As such, the analysis of data focused on describing patient satisfaction, patient fall, and HAPU rates at the beginning and end of the study period. With respect to the time of data collection, the study was prospective in nature. A prospective study involves studying outcomes during the study period with the aim of establishing the impact of a given intervention (Maltby et al., 2015). In this case, satisfaction scores as well as fall and HAPU rates were measured before, during, and after the implementation of the nurse rounding intervention.

Though descriptive research provides information about the attributes of a given research phenomenon, it provides little or no knowledge about cause-and-effect relationships (Bryman, 2008). In this case, for instance, it may not be said with certainty that hourly nurse rounding actually increased patient satisfaction as patient satisfaction may be influenced by other factors within the care environment. Further, lack of randomization and a control group may present generalization difficulties.

Sampling

As mentioned earlier, the study was carried out in a 24-bed medical-surgical unit in a large hospital. In total, 582 eligible patients were discharged during the study period. However, only 81 patient surveys were returned. Given the nature of the study, the recruitment of the sample involved non-random techniques. In other words, outcomes were observed in all patients that got admitted to the unit during the study period. Lack of random sampling presents a threat to internal and external validity due to the possibility of researcher bias (Maltby et al., 2015). Furthermore, the final sample (n = 81) is fairly small in view of the larger population under investigation, making it not sufficiently representative of the population.

Measurement Methods

Data collection is one of the most important elements of the research process. Accordingly, data collection procedures must be undertaken with a great deal of caution. One of the ways of ensuring effective data collection is ensuring data collectors have extensive knowledge of what data they are supposed to collect and how to collect it (Maltby et al., 2015). In Brosey & March's (2015) study, the data collection process was preceded by a 20-minute education session aimed at familiarizing the unit's staff (the nurse manager, registered nurses, patient care assistants, and secretaries) with related evidence, the intervention, historical, performance indicators, and goals of the intervention. Training was crucial since the unit's staff would be the data collectors.

Effective data collection also entails choosing an appropriate data measurement tool. When it comes to data collection, a researcher may choose between a researcher-designed tool and standardized tool (Creswell, 2014). Compared to a researcher-designed tool, a standardized tool tends to be more effective as it has been scientifically tested for validity and reliability (Maltby et al., 2015). In this case, patient satisfaction was measured using the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey. The HCAHPS survey is a nationally applicable tool for measuring patient satisfaction with healthcare, thereby an appropriate tool for the study. This enhances validity and reliability.

Descriptive and Inferential Statistics

The analysis of the data mainly involved descriptive statistics. This was primarily informed by the descriptive nature of the study. Descriptive statistics basically involves describing the attributes of data using techniques such as frequencies, tables, and graphs (Martin & Bridgmon, 2012). Being a descriptive study, inferential statistics were not employed. For instance, elements such as effect sizes and confidence intervals were not used. Also, as the study did not involve hypotheses, type I and type II errors are not relevant. Nonetheless, a Cox-Stuart trend analysis was performed on data collected 12 months before the implementation of the intervention. The analysis was conducted to determine the statistical significance of reduction in patient falls in the 12-month period preceding the intervention.

Data Management

Data management is an important aspect of the data analysis process (Broeck et al., 2005). It involves processes such as cleaning and coding data, checking for outliers and nature of data distribution, handling missing data, as well as inter-rater reliability. In Brosey & March's (2015) article, it has not been mentioned whether any procedures were undertaken to clean the collected data prior to analysis, to deal with incomplete responses in the HCAHPS questionnaires, or to determine the distribution of the data. Inter-rater judgment was also not done. The only aspect reported is attrition: out of the 582 eligible patients, only 81 returned their HCAHPS surveys. The absence of most data management aspects can be viewed as one of the major weaknesses of the data analysis process in this study.

Mitchell, M., Lavenberg, J., Trotta, R., & Umscheid, C. (2014). Hourly rounding to improve nursing responsiveness. A systematic review. The Journal of Nursing Administration, 44(9), 462-472.

Research Purpose

This is a systematic review aimed at synthesizing empirical evidence on the impact of hourly nurse rounding on patient satisfaction. As this was not a primary study, no hypothesis was formulated.

Research Design

A systematic review is a type of research method aimed at summarizing and critically analyzing the available evidence on a given research topic (Garg, Hackam & Tonelli, 2008). This way conclusions or inferences are drawn from numerous studies as opposed to a single study. For studies within the field of healthcare, systematic reviews are valuable as readers can readily establish the effectiveness of a given intervention. Indeed, systematic reviews are often described as the most robust form of medical evidence (Maltby et al., 2015). This is one of the major advantages of a systematic review. Furthermore, systematic reviews involve less time and costs, and have better generalizability. Nonetheless, the quality of a systematic review is significantly dependent on the nature of studies included in the review in terms of date of publication, type of research design, measurement tools used, and outcomes measured (Maltby et al., 2015). These aspects determine the reliability of a systematic review. For instance, a review that includes only randomized-controlled trials provides stronger evidence than one that includes descriptive studies.

In this case, the studies included were controlled clinical studies. No randomized controlled trial was included, which could be viewed as one of the major limitations of the review. The studies included reported patient satisfaction outcomes as a result of the implementation of hourly nursing rounds for a period ranging from two weeks to 5 years. More importantly, all the studies included had been published within 6 years as at the time of completing the review. This adds to the strengths of the review given the significance of presenting up-to-date evidence in healthcare research (Maltby et al., 2015). As this was a systematic review, the timing of data collection or subject enrolment is not relevant.

Sampling

CINAHL, EMBASE, and Medline databases were used as the sources for the articles. Random selection was involved in locating the articles. In the first stage, an expert was engaged to evaluate the titles and abstracts of all the articles located through searches in the three databases. The analyst marked articles to be considered in the second stage. Out of the articles located in the first stage, 100 articles were selected randomly. A second analyst then evaluated the 100 articles as per the inclusion criteria. The evaluation also involved checking for possible bias. These procedures are important for improving the reliability of the findings of the review. Eventually, 16 articles were found to meet the inclusion criteria. This is a fairly sufficient number of articles for a systematic review, thereby reinforcing the generalizability of the findings.

Measurement Methods

For a systematic review, the data collection process involves locating the studies to be included in the review. This means that the measurement methods used differ from those used in primary studies. Measurement within the context of a systematic review entails checking the quality of potential studies. In this regard, the researchers relied on two tools: the Jadad scale and the Grading of Recommendations Assessment, Development and Evaluation (GRADE) scale. With rigorous testing, these scales provide useful guidelines for evaluating the quality of a study or evidence with regard to aspects such as research design, consistency, and magnitude of effect. This has important implications for validity and reliability. As this was not a primary study, data collector training was not relevant. Nonetheless, most of the studies included in the review included personnel training in the data collection process.

Descriptive and Inferential Statistics

The review reports descriptive statistics. For instance, descriptive techniques were employed to describe the characteristics of the studies included, particularly in terms of number of subjects, duration of study, frequency of nursing rounds, rate of patient falls, use of call lights, nurse responsiveness, and patient satisfaction levels. Nonetheless, there is little use of inferential statistics. The only type of inferential statistics reported is correlation (association between hourly nursing rounds and patient satisfaction). The little reporting of inferential statistics may be attributed to the character of the underlying studies included in the review. Effect sizes and statistical significance were reported. This was important for showing the strength of the evidence. Confidential intervals were not reported. Since this was a systematic review, type I and type II errors are not relevant.

Data Management

Data management in a systematic review involves different procedures compared to primary studies. In a systematic review, data management essentially encompasses evaluating the strength of evidence. In this case, this was achieved by using two experts. The experts thoroughly evaluated the titles, abstracts, and contents of all the articles. The evaluation involved assessing research methodology and potential bias. The use of two experts can be viewed as one of the major strengths of the data management process. Each expert did an independent evaluation. Results were then compared to establish agreements and disagreements. Little disagreements between the experts were found.

Singh, S., Fletcher, K., Pandl, G., Schapira, M., Nattinger, A., Biblo, L., & Whittle, J. (2011). It's the writing on the wall: whiteboards improve inpatient satisfaction with provide communication. American Journal of Medical Quality, 26(2), 127-131.

Research Purpose

The purpose of this study was to examine the effect of whiteboards on inpatient satisfaction with nurse or physician communication. The boards were placed on medicine wards in each patient room in an effort to fulfill the informational needs of patients, particularly in terms of medication and discharge instructions.

Research Design

Dissimilar to Brosey & March's (2015) study, this study can be described as an experimental study. The study was carried out to compare changes in patient satisfaction in two groups: the intervention group and the control group. In the intervention group (patients in medicine wards), whiteboards containing discharge and medication information were placed in each patient room. The control group involved patients in surgical wards. For the control group, whiteboards were not placed on wards. The study was prospective in nature in the sense that patient satisfaction outcomes were measured before and after the placement of the whiteboards.

Experimental research is the gold standard of scientific research as it determines cause-and-effect relationships (Creswell, 2014). This is unlike descriptive research or correlational research, which provide a description of study variables and the association between them. Nonetheless, association does not necessarily imply causation. Experimental research overcomes this challenge. In this case, for instance, it can be said with certainty that the improvement in patient satisfaction in the intervention group was as a result of the use of whiteboards. A major limitation of the study, however, is that there was no randomization -- quasi-experimental research. An experiment is stronger when subjects are randomly assigned to groups as this often eliminates researcher and allocation bias (Maltby et al., 2015). Even so, randomization may not have been possible or necessary in this case given the purpose of the study.

Sampling

The study was carried out at a 430-bed teaching medical center located in an urban area in Midwestern U.S. Approximately 37% of the patients discharged from each ward were emailed the survey. The recipients of the survey were selected randomly. Use of random sampling is one of the major strengths of this study. Random sampling means every element of the study population has an equal chance of inclusion, thereby minimizing or avoiding selection bias (Bryman, 2008). This is vital for enhancing validity. The sample used (37% of the total discharged patients) is another strength of the study. This percentage is fairly substantial, increasing the generalizability of the findings to the larger population.

Measurement Methods

The major variable examined in this study was patient satisfaction. The variable was measured using the Press Ganey Patient Satisfaction Survey, a standardized commercial tool for measuring patient satisfaction. The instrument is utilized in over 7,000 healthcare facilities, a clear indication of its popularity. This eliminates doubts about the validity and reliability of the tool. The tool requires respondents to rate their satisfaction on a 5-point Likert scale. It is not clear from the article whether data collector training was conducted. Nonetheless, physicians received email notifications from the clinical leader reminding them of the initiative.

Descriptive and Inferential Statistics

Both descriptive and inferential statistics were used. Descriptive statistics mainly included measures of central tendency and dispersion such as the mean and standard deviation of patient satisfaction scores. Inferential statistics mainly involved t-tests. The tests were done to compare patient satisfaction scores before and after the installation of white boards. It emerged that patient satisfaction scores in the intervention group had increased significantly following the placement of whiteboards.

Confidence intervals and effect sizes were not reported, which could be seen as a weakness of the data analysis process in this study. It would have been valuable to inform the reader about the uncertainty associated with the sample as well as the practical significance of the findings. However, statistical significance was reported. The resultant p values show that the whiteboards significantly improved patient satisfaction in the medicine wards. Since the researchers did not specify a hypothesis, type I and type II errors are quite irrelevant.

Data Management

There is little information about data management processes from the article. The only two aspects reported are attrition and the shape of data distribution. Of all the emailed surveys, only 28% were returned. This rate is quite comparable to the HCAHPS rate. Data distribution was checked by determining standard deviation. It would have been important for the researchers to reveal whether there were any instances of missing data. Also, inter-rater judgment would have strengthened the reliability of the findings.

Tan, M., Evans, K., Braddock, C., & Shieh, L. (2013). Patient whiteboards to improve patient-centered care in the hospital. Postgraduate Medical Journal, 89, 604-609.

Research Purpose

The purpose of this study was to examine the impact of whiteboards on patient-centered care. The study specifically sought to achieve three objectives: 1) to determine the impact of whiteboards on patient's understanding of and satisfaction with care; 2) to understand factors that hinder physicians from using whiteboards; and 3) to explore the impact of whiteboards on the families of patients.

Research Design

Similar to Singh et al.'s (2011) study, this study was also carried out as an experiment. This is one of the major strengths of the study. Further, the experiment in this case involved randomization, making the study more rigorous. Conducted as a pilot project, the study was implemented in four inpatient units. Two of the units were used as the intervention group (whiteboards were placed in patient rooms) and the other two as the control group (whiteboards were not placed in patient rooms). Based on the admission call schedule, patients were then randomly allocated to general medicine teams in the selected units. As mentioned earlier, random allocation in experimental research overcomes the challenge of selection bias. Conducted over a period of three weeks, the study was prospective in nature.

Sampling

The study was conducted at Stanford University's Medical Center. The hospital is a 613-bed facility, admitting approximately 25,000 patients annually and with more than 2,100 medical staff. As this study aimed to determine the impact of whiteboards on patient satisfaction, to understand factors that hinder physicians from using whiteboards, and to explore the impact of whiteboards on the families of patients, subjects included patients, staff, and patients' families. This is one of the major differences between this study and others in this critique, which collected data only from patients.

In total, 104 patients participated in the study: 56 from units in the intervention group and 48 from units in the control group. For medical staff, 25 residents participated in the study. For both patient and medical staff, no random sampling techniques were employed. Patients were recruited on the basis of the following inclusion criteria: aged 18-91 years, admitted in general medicine and for a non-surgical condition, and admitted for at least three days. A convenience sample comprising 8 family members, 8 occupation therapists, 8 physical therapists, 8 case managers, and 8 consultants was also recruited. Convenience sampling is a form of non-probability sampling that involves recruiting subjects on the basis of their accessibility and proximity to the researcher (Bryman, 2008).

On the whole, the use of non-random sampling can be viewed as one of the major weaknesses of this study. Non-random sampling can often result in selection bias, presenting a challenge to validity. Even so, the sampling method was quite appropriate given the objectives of the study. As for sample representativeness, the sample used is fairly small compared to the population under study. This limits the extent to which the findings of the study may be generalized beyond the study setting.

Measurement Methods

There are two important aspects in measurement methods: data collection tools and data collector training. A researcher-designed survey tool was used to collect data from patients. Based on the input of Stanford's Quality Improvement Team, the questionnaire was a 5-point Likert scale aimed at measuring patients' knowledge of aspects such as the physician's name, reason for hospitalization, plan of care, and expected date of discharge. Though the tool was developed by experts, the challenge of reliability and validity cannot be ignored. In other words, the tool has not been standardized or scientifically tested, limiting its use in other settings. Nevertheless, to ensure accurate filling, the principal researcher delivered the questionnaire to patients in person and completed it together with the patient. Data from medical staff was obtained using an electronic audience polling device. It is not clear from the article whether data collector training was done. However, all members of staff in the selected units were trained on updating the whiteboards.

Descriptive and Inferential Statistics

Both descriptive and inferential statistics were used. Descriptive statistics were used to summarize patients' responses as well as challenges faced by medical staff in using whiteboards. This mainly involved measures of central tendency and graphs. Being an experimental study, the employment of inferential statistics was important. More specifically, the researchers conducted a rank sum test for patient responses, and compared scores between the two groups. The choice of a rank sum test, instead of a t-test, was informed by the nature of the data (non-parametric). A multivariate analysis of variance (MANOVA) was conducted to account for variance. Further, a power analysis was conducted using effect size. A Monte Carlo simulation was carried out to confirm the results. The results were statistically significant meaning that patients in wards with whiteboards were significantly more satisfied than those in wards without whiteboards. From the article, it is not clear whether confidence intervals were used. Moreover, as no hypothesis was formulated, type and I type II errors may not be relevant. On the whole, this study had stronger data analysis techniques compared to the rest in this critique.

Data Management

This study also depicts better management of data compared to the rest. First, it is clearly indicated that the data was heavily skewed to the left. This means that the researchers checked for the shape of distribution of data. Attrition has also been reported. Out of all eligible patients, none declined to respond to the questions, meaning a 100% response rate. This is a major strength of the study as it is quite rare to achieve a 100% response rate. As patients filled the questionnaire together with the principal investigator, it is unlikely that there were cases of missing data or incomplete responses. Nonetheless, similar to most studies in this critique, data cleaning and inter-rate judgment was not employed.