Nursing Diagnosis Risk for Unstable Blood Glucose Level in Patients with Diabetes Mellitus

Introduction

Diabetes mellitus (DM) is a disease high prevalence in the world. Currently, it is estimated that the world's population with diabetes will reach 471 million in 2035 and is thus a world health problem. An unstable blood glucose level can result in the long-term damage of blood vessels, nerves, and other tissues. Besides that, hyperglycemia combined with other risk factors (dyslipidemia, obesity, smoking, and physical inactivity) can cause complications, such as cardiovascular diseases, retinopathy, nephropathy, and nervous system dysfunction (European Coalition for Diabetes, 2014).

Patients with type 2 diabetes and increased glycated hemoglobin have greater risk of developing macrovascular changes (heart failure and cardiovascular disease) and microvascular abnormalities (lower limb amputations, blindness, and renal failure) (Cox & Coupland, 2016).

In view of the complications of unstable blood glucose level in patients with DM, it is necessary to identify risk factors for unstable blood glucose and to develop preventive measures. Nurses, working in different scenarios and health care levels (primary, secondary, and tertiary), play a leading role in the implementation of systematic nursing care with a focus on preventing complications of chronic diseases, and above all, in the pursuit of patient adherence to diabetes management plan that is necessary to an effective glycemic control (Castro-Sampaio et al., 2017).

Risk diagnoses represent a clinical judgment about the vulnerability of an individual, family, and community, to develop undesirable human responses to health conditions (Herdman & Kamitsuru, 2017). This kind of diagnosis provides the basis to select nursing interventions in order to achieve satisfactory outcomes, for which the nurse is accountable (Cavalcante et al., 2013).

To further enhance the taxonomy of nursing diagnoses, validation and accuracy studies are needed. In addition, the latest version of the NANDA-I classification (Herdman & Kamitsuru, 2017) shows the inclusion of new categories: at risk population and associated condition. According to the authors, these new categories were developed by repositioning some related factors or risk factors, and thus it is necessary to carry out new studies.

The submission of new diagnoses or the review of existing diagnoses strengthens the breadth and evidence supporting the NANDA-I. The developers of this taxonomy value studies aiming at validating and improving its diagnoses in specific populations and contexts in order to increasingly improve the taxonomy, by the recognition of altered human responses, thereby contributing to a more effective nursing care (Herdman & Kamitsuru, 2017).

Teixeira, Tsukamoto, Lopes, and Silva (2017) found in their study risk factors of unstable blood glucose level in people with type 2 diabetes that are not listed in the current version of the NANDA-I classification, such as age, black race, delayed diagnosis of diabetes, daytime sleepiness, macroalbuminuria, genetic polymorphisms, use of insulin, use of oral antidiabetic drugs, use of metoclopramide, inadequate physical activity, and low fasting glucose.

In the literature, few studies have focused on the nursing diagnosis risk for unstable blood glucose level in people with DM. Some studies (Moura et al., 2014; Teixeira et al., 2017) about the prevalence of this nursing diagnosis in patients with hypertension and DM were developed; however, we did not find validation studies of this diagnosis in adults or the elderly patients with type 1 and type 2 diabetes.

Due to the epidemiological importance of DM, the need to improve the taxonomies of nursing diagnoses, and the relevance of knowledge about risk factors for unstable blood glucose in clinical practice, this study aims to validate the nursing diagnosis risk for unstable blood glucose level in adult/elderly patients with type 1 and type 2 diabetes.

Method

A methodological study was conducted. According to Polit and Beck (2006), this type of research is relevant when authors aim to create, validate, and assess tools and techniques for research and clinical practice. The validation process followed strictly the steps proposed by Hoskins (1989), which consists of: (a) concept analysis, (b) validation by experts, and (c) clinical validation. In this study, only the first two steps were developed.

The concept analysis was performed using Walker and Avant's (2005) method, which covers eight steps: (1) concept selection, (2) determination of aims, (3) identification of uses of context, (4) determination of defining or essential attributes of context, (5) identification/construction of a model case of context, (6) identification/construction of additional cases of context (borderline, related, and contrary cases), (7) identification/construction of antecedents and consequents of context, and (8) definition of empirical referents of context.

In order to support the implementation of the concept analysis, an integrative literature review was conducted, covering five steps: problem identification, literature discovery, data evaluation focused on methodological quality, data analysis, and presentation of results (Whittemore & Knafl, 2005).

The research questions of the integrative review, that supported the concept analysis, were: “What are the essential attributes, the antecedents, and the consequents of unstable blood glucose?”; “What are the risk factors, at risk populations, and associated conditions of unstable blood glucose in adult/elderly patients with type 1 and type 2 diabetes?”; “How the risk factors for unstable blood glucose are measured in adult/elderly patients with type 1 and type 2 diabetes?”

The search was conducted in the following databases: LILACS, BDENF, PubMed, Scopus, Web of Science, and Cochrane, using the controlled descriptors: diabetes mellitus, risk factor, hypoglycemia, and hyperglycemia combined by the operator “AND.” The search was not limited by time in order to achieve a maximum number of studies for the conceptual analysis.

Inclusion criteria were: articles in Portuguese, English, or Spanish; with full-text available; and that focused on risk factors, at risk populations, or associated conditions for unstable blood glucose level in adult and elderly patients with type 1 and type 2 diabetes.

The selected studies were classified using the levels of evidence described by Melnyk and Fineout-Overholt (2005). The analysis and presentation of results were made by dividing the selected studies per database used, and highlighting, in a frame, the main evidences from each article (concepts, antecedents, consequents, at risk population, and associated condition of unstable blood glucose).

Based on the NANDA-I taxonomy and results from the concept analysis, an instrument with the concept, risk factors, empirical referents for risk factors, at risk population, and associated condition of the nursing diagnosis risk for unstable blood glucose level was created for subsequent validation by experts.

Experts were selected using a Fehring's (1994) adapted model. The sample size was defined by n = Z _α ².P.(1–P)/d ², where “Z _α” refers to the confidence level (we adopted a confidence level of 95%), “P” is the proportion of experts who agree on the relevance of the nursing diagnosis component, and “d” is the acceptable difference in proportion. The final calculation was defined by n = 1,96².0,85.0,15/0,15², resulting in 22 experts (Lopes, Silva, & Araújo, 2013).

Experts were recruited using three strategies: (a) by searching researchers’ CVs available at the Lattes Platform of the National Scientific and Technological Development Council (CNPq, Brazil) using keywords, such as: risk for unstable blood glucose level, type 1 and type 2 diabetes, and nursing diagnoses; (b) by personal recruitment in strict sense graduate programs in nursing and medicine; and (c) by interpersonal relations and connections between researchers selected using the first two strategies (snowball sampling).

Data were compiled into a Microsoft Excel 2010 spreadsheet and analyzed using the Statistical Package for the Social Sciences (SPSS) version 20.0 software.

To check the appropriateness of adjusting the proportions of experts who agreed on the relevance of each item evaluated (appropriateness of the concepts, risk factors, empirical referents for risk factors, at risk population, and associated condition of risk for unstable blood glucose level), items were grouped dichotomously, by joining the scores 1–3 of the Likert scale into the “inappropriate” category and joining the scores 4 and 5 into the “appropriate” category.

The binomial test was used at this stage to adjust the proportion of agreement, considering a proportion of 85% concordant experts. For the significance analysis, a significance level (p) of 5% was considered to reject the null hypothesis.

The project was submitted to the Ethics Committee of the University of International Integration of the Afro-Brazilian Lusophony (UNILAB––protocol number 2522734). All study subjects were invited to participate in the survey, received a letter of invitation, and signed an informed consent form.

Results

After the literature search, a total of 277 publications were selected (Table 1), composing the final sample of studies included in the concept analysis of unstable blood glucose in adult/elderly patients with type 1 and type 2 diabetes.

Regarding the assessment of the methodological quality of the selected studies, prevailed authorities’ opinion studies and expert committees’ reports (39.41%), followed by descriptive or qualitative studies (28.88%), systematic reviews with meta-analysis (14.79%), randomized controlled trials (10.1%), cohort and case–control studies (4.69%), and systematic and literature reviews of descriptive and qualitative studies (2.52%). Most of the studies were conducted in the United States (37.18%) and the UK (5.77%).

The concept of unstable blood glucose was found in seven publications. After the concept analysis, essential attributes were identified: variability, fluctuations, glucose, and blood. These essential attributes are present in the definition of the nursing diagnosis risk for unstable blood glucose level as shown in the NANDA-I taxonomy.

A total of 14 risk factors (antecedents) of unstable blood glucose in adults/elderly patients with diabetes were found in the studies, as follows: fasting, insufficient dietary intake, ineffective medication management, polypharmacy, high carbohydrate and lipid food intake, excessive weight gain, average daily physical activity is less than recommended for gender and age, excessive stress, daytime sleepiness, unknown hypoglycemia, fear of hypoglycemia, nonadherence to diabetes management plan, long-distance travel, and use of medicinal plants.

By comparing the 14 risk factors identified with those presented by the NANDA-I taxonomy, eight risk factors did not overlap with the original ones: fasting, polypharmacy, high carbohydrate and lipid food intake, daytime sleepiness, unknown hypoglycemia, fear of hypoglycemia, long-distance travel and use of medicinal plants.

We also identified new at risk populations and associated conditions of risk for unstable blood glucose level in patients with type 1 and type 2 diabetes, as follows: at risk population: elderly, blacks, current smokers, patients with comorbidities, genetic predisposition, unfavorable social conditions, lack of access to health services and difficulties in performing daily living activities, lack of social support, and cognitive impairment; associated condition: dehydration, electrolyte imbalance, malnutrition, overweight and obesity, use of oral antidiabetic drugs, use of immunosuppressants, combined drug therapy, enteral and parenteral nutrition, and surgery.

In the second stage of the study (validation by experts), most experts were female (77.3%), with a doctors’ degree (63.6%), and with developed doctoral thesis involving nursing taxonomies (50%). Regarding research activities, most said that they currently participate in research groups on nursing diagnoses (68.2%), develop scientific articles (81.8%), and provide mentorship on studies (90.9%) about nursing diagnoses and unstable blood glucose (77.3%).

Regarding the identification of risk for unstable blood glucose level in the experts’ clinical practice, 63.6% said that they often identify this diagnosis, and 86.4% reported that they use nursing diagnoses frequently in their practice and have already provided care for patients with diabetes and unstable blood glucose level.

Regarding the experts’ evaluation of the relevance of the identified risk factors, the risk factors “ineffective medication management” and “high carbohydrate and lipid food intake” achieved a p value <.05 in the binomial test, and experts’ agreement ratio greater than 85%, thus rejecting the established hypothesis. Some risk factors achieved an experts’ agreement ratio significantly below 85% (p<.05), which indicates the need for revision or exclusion in the future (Table 2).

The experts also suggested the inclusion of new risk factors: lack of glucose control, insufficient knowledge of diabetes, unfavorable socioeconomic conditions, low education, and visual problems. Regarding the at risk population found during the concept analysis, “blacks” and “current smokers” achieved an experts’ agreement ratio inferior to 85% (p values .001 and .037, respectively), and therefore were excluded from the final instrument.

There was not a significant proportion of agreement among the experts about the associated conditions of risk for unstable blood glucose level in patients with DM, considering p<.,05, and all remained in the instrument. Three experts have suggested the inclusion of new associated conditions: polycystic ovary syndrome and insomnia.

After the evaluation of risk factors, empirical referents, at risk populations, and associated conditions by experts, a new proposal for the nursing diagnosis risk for unstable blood glucose level in adult and elderly patients with diabetes was developed, obtaining the final product described in Table 3.

Discussion

Seven studies presented the concept of unstable blood glucose (Akirov, Diken-Cohen, Masri-Iragi, & Shimon, 2017; Cavalot, 2013; Deane & Horowitz, 2013; Fabris et al., 2014; Hirsch et al., 2012; Sertbas et al., 2017; Teixeira et al., 2017). Glycemic variability is a term used for diabetic patients who have similar mean glucose levels in varying degrees, leading to episodes of hyperglycemia and hypoglycaemia (Sertbas et al., 2017).

Ineffective medication management doses and omission of doses trigger changes in blood glucose levels that can lead to complications, such as diabetic ketoacidosis and hipeosmolar hyperglycemic state. Reasons for neglected insulin doses comprise low adherence to the regimen, psychological factors, and limited financial resources (Pollock & Donna, 2013). Difficulty in adhering to diet, eating fried foods and carbohydrates, and improper use of medications can also lead to changes in glycemic values (Pinelli & Jaber, 2011).

Some risk factors were judged as inadequate by the experts: polypharmacy, excessive weight gain, average daily physical activity is less than recommended for gender and age, excessive stress, daytime sleepiness, fear of hypoglycemia, unknown hypoglycemia, long-distance travel, and use of medicinal plants (Table 1). The risk factors excessive weight gain, excessive stress, and average daily physical activity is less than recommended for gender and age did not present a satisfactory significance; however, they were maintained as they already belong to the diagnosis risk for unstable blood glucose level (Herdman & Kamitsuru, 2017). The risk factors long-distance travel and use of medicinal plants were also maintained.

According to Pinsker et al. (2017), long-distance travel is a risk factor for unstable blood glucose in people with type 1 diabetes due to lack of information, presence of unfamiliar foods, time zone changes, inadequate insulin storage during the travel, among others. In addition, plants with mangiferin content, African walnut, kola biter, bitter leaf, cinnamon, turmeric, scent leaf, carob, and ginger can trigger glycemic changes (hypoglycemia and hyperglycemia) in people with diabetes (Ogunyinka, Oyinloye, Adenowo, & Kappo, 2015).

Regarding the suggestion of adding the risk factor “visual problems,” Carvalho, Freitas, Araújo, Zanetti, and Damasceno (2017) point out that, in addition to the lack of knowledge about the disease and skills needed for the management of insulin doses, visual deficits can result in excessive or insufficient doses. Thus, this item was included as an “at risk population” of the nursing diagnosis risk for unstable blood glucose level in patients with DM.

Regarding blood glucose monitoring, patients with type 1 and type 2 diabetes in insulin use are recommended to monitor their blood glucose level three or more times a day (Brazil Ministry of Health, 2013). However, while recognizing that the monitoring of blood glucose is a key part of the treatment, it is known that many patients, for various reasons (economic, social, or psychological), present difficulties in carrying it out (Franco, Zanetti, Teixeira, & Kusumota, 2008). This risk factor is presented in the latest version of the NANDA-I taxonomy (Herdman & Kamitsuru, 2017), therefore, we decided to keep it in the proposed diagnostic structure.

Regarding the suggestion of including the associated conditions “polycystic ovary syndrome” and “insomnia,” Pereira, Silva, and Cavalcanti (2015) show that despite polycystic ovary syndrome favors the occurrence of various diseases, including type 2 diabetes, high cholesterol, excessive weight gain, and high blood pressure, there is still no consensus on the development of insulin resistance.

As for insomnia, a study on the profile of nursing diagnoses in patients with hypertension and diabetes showed that insomnia was present in 51.4% of the sample (Castro-Sampaio et al., 2017). In addition, difficulty in maintaining sleep in diabetic patients may mean more than fatigue, since it interferes directly with metabolic control, production of glucocorticoids, and glucose levels (Cunha, Zanetti, & Hass, 2008).

Thus, it is believed that the findings of this research have appropriate content to support the identification of risk factors for risk for unstable blood glucose level in adult/elderly patients with DM, as well as its associated conditions and at risk populations.

Conclusion

The concept of analysis was a valuable step for the development of this work, particularly because it allowed us to explore more consistently the attributes of the central concept of this study (unstable blood glucose), providing a greater contribution to the development of the concept and the identification of important new risk factors for people with DM not yet contemplated in NANDA-I taxonomy, such as fasting, high carbohydrate and lipid food intake, long-distance travel, and use of medicinal plants, as well as the building of empirical referents based on the investigation of the ways of measuring the antecedents (risk factors) of unstable blood glucose in patients with diabetes.

It is important to note that among the risk factors found in this study, most overlapped with the ones that are already described in the NANDA-I taxonomy, and thus have been kept: insufficient dietary intake, ineffective medication management, excessive weight gain, average daily physical activity is less than recommended for gender and age, excessive stress, nonadherence to diabetes management plan, inadequate blood glucose monitoring, and insufficient knowledge of disease management.

Regarding the at risk populations identified in the concept analysis, only nine remained in the final version. The risk factor visual problems have not been submitted for expert evaluation, since it was included after suggestions from them. Besides that, “insomnia” was added as an associated condition as suggested by the experts.

Given these findings, it is believed that the new risk factors, at risk populations, and important associated conditions identified for the nursing diagnosis risk for unstable blood glucose level can be used to improve clinical protocols for the prevention of unstable blood glucose, in addition to contributions to the advancement of nursing science and to the improvement of classification systems.

We also believe that this research can contribute to the development of other studies aimed at clinical validation of nursing diagnosis risk of unstable blood glucose.