Quality healthcare should be grounded on clinical practice with the highest benefit-risk ratio and cost-effectiveness according to the available scientific evidence. The overuse of unproven diagnostic or therapeutic procedures is common in our setting and leads to increased healthcare spending and even iatrogenic harm. Previous cost-effectiveness initiatives have proposed identifying diagnostic and therapeutic measures that are better 'not done' in certain clinical contexts under the lens of the available scientific evidence. In this regard, the Spanish Society of Endocrinology and Nutrition (SEEN) has compiled a series of 'not-to-do' recommendations from its various working groups. These recommendations cover common clinical situations classified into the following thematic areas: diabetes, nutrition, pituitary gland, neuroendocrine tumors, thyroid, and hormone replacement therapy in postmenopausal women.
La asistencia sanitaria de calidad es aquella que se basa en las prácticas clínicas que tienen el mayor cociente beneficio-riesgo según la evidencia científica disponible y a unos costes razonables. La sobreutilización de procedimientos diagnósticos o terapéuticos de eficacia no probada es común en nuestro medio y genera un incremento del gasto sanitario y, potencialmente, yatrogenia. Iniciativas previas de coste-efectividad han propuesto identificar medidas diagnósticas y terapéuticas que sea mejor “no hacer” en determinados contextos clínicos bajo el prisma de la evidencia científica disponible. En esta línea, la Sociedad Española de Endocrinología y Nutrición (SEEN) ha recogido una serie de recomendaciones de no hacer emanadas de sus diferentes grupos de trabajo y áreas de conocimiento. Estas recomendaciones cubren varias situaciones clínicas comunes clasificadas por las siguientes áreas temáticas: diabetes, obesidad, nutrición, hipófisis, tumores neuroendocrinos, tiroides y terapia hormonal sustitutiva en la mujer postmenopáusica.
The overuse of diagnostic or therapeutic procedures with unproven efficacy is common in our setting and leads to increased health care costs.1 Overdiagnosis can occur due to screening tests in healthy populations without scientific evidence to support them or by requesting diagnostic tests in patients already diagnosed with a condition, which only increases risk, discomfort, and cost.2 Overtreatment arises when unnecessary, harmful, or non-beneficial therapeutic procedures are used, providing no added value.3
Quality health care is based on clinical practices with the highest benefit-risk ratio according to available scientific evidence and at reasonable costs.4,5 Various initiatives have been undertaken to identify low-value interventions and improve appropriateness.6,7 In 2012, the American Board of Internal Medicine (ABIM) Foundation launched the Choosing Wisely campaign,8 establishing recommendations from nine societies, which have since expanded to over 500. In Spain, the Ministry of Health's Quality Plan for the National Health System9 aims to document and propose initiatives to reduce unjustified variability in clinical practice and promote the development and use of Clinical Practice Guidelines linked to Health Strategies. The GuíaSalud project10 is responsible for the methodological coordination of this initiative. Essencial,11 an initiative by the Agència de Qualitat i Avaluació Sanitàries de Catalunya (AQuAS), identifies low-value practices and promotes recommendations to avoid them. The DianaSalud portal, developed by the Clinical Epidemiology Program of the Biomedical Research Network in Epidemiology and Public Health (CIBERESP), coordinated by the Epidemiology and Public Health Service of Hospital de la Santa Creu i Sant Pau in Barcelona (Catalonia, Spain),12 allows consultation of national and international initiatives to identify low-value interventions and improve appropriateness.13 The MAPAC initiative (Intitiative to Improve Health and Clinical Care Practices), through a defined methodology, aims to systematically analyze the appropriateness of clinical and health care services and formulate specific recommendations and actions for improvement.
In line with these initiatives, the Spanish Society of Endocrinology and Nutrition (SEEN) proposed in 2023 to draft a document compiling a series of "do not do" recommendations from its various working groups (WG) and knowledge areas (KA), aiming to identify inappropriate or low-value attitudes, services, or clinical procedures within our specialty.
MethodsThe present project was approved by the SEEN board on September 28th, 2023. All KA and WG were invited to participate, with acceptance from the 5 KAs (diabetes, obesity, nutrition, neuroendocrinology, and thyroid) and the gonads, identity, and sexual differentiation working group. Coordinators of each KA and WG appointed a project leader responsible for selecting the health problem and health care procedure for review within their area of interest, as well as reviewing the literature and drafting relevant recommendations.
In selecting the health problem and procedure, authors considered clinical relevance based on the prevalence of the health issue and the severity of the drawbacks of inappropriate practice. Once selected, the WG and KA leaders conducted a literature review to identify evidence supporting the recommendations. Priority was given to synthesis documents, clinical practice guidelines, systematic reviews, and randomized controlled trials. Once the bibliographic sources were identified, it was assessed whether they were suitable for the selected problem or procedure, that is, whether the patients and the intervention for which the recommendation was to be formulated were properly included in the selected references.
Each recommendation was developed in a structured manner in three sections: writing, source of evidence, and justification. The writing was done in a clear, direct, and specific manner, in a single paragraph. Subsequently, justificatory paragraphs were included, providing a brief explanation of the factors that determined the selection of the recommendation, as well as a summary justifying the available evidence. Potential benefits for patients and the healthcare system were also included.
After the drafting of the recommendations by the responsible person, all members of each WG and KA performed an initial review, providing their suggestions or comments. Once the final draft of the recommendations was completed by the responsible person of each WG and KA, the coordinators of the WG of the SEEN board compiled and organized the recommendations, unified the style, and prepared the document final draft. After the compilation, the document was returned to all the responsible individuals of each participating WG and KA for a second review. After this second review, the coordinators revised the final version with the received comments and suggestions. Subsequently, the final document was submitted to a third review by all the members of the board of directors of SEEN.
RecommendationsDiabetesRecommendation #1Do not use metformin universally to prevent diabetes in individuals with prediabetes.
Evidence sourceMeta-analyses, systematic reviews, and randomized controlled trials.14–26
JustificationThree arguments support this recommendation: (a) approximately two-thirds of individuals with prediabetes do not develop diabetes, even after many years; (b) about one-third return to normal glucose regulation; and (c) individuals meeting prediabetes criteria have a very low risk of diabetes-related microvascular complications, making the impact of metformin minimal.
Recommendation #2Metformin is not recommended for treating gestational diabetes.
Evidence sourceMeta-analyses, systematic reviews, and randomized controlled trials.27–34
JustificationInsulin, not metformin, is recommended as first-line therapy since metformin crossing the placenta, raising safety concerns. Metformin should be discontinued by the end of the first trimester if used for polycystic ovary syndrome. While metformin is associated with lower neonatal hypoglycemia and maternal weight gain vs insulin, it easily crosses the placenta, resulting in fetal exposure. Controlled trials and meta-analyses have shown growth alterations in fetuses exposed to metformin, including lower birth weight and higher BMI during childhood and adolescence.
ObesityRecommendation #1Do not use prescription or over-the-counter drugs for weight control that have not been approved by government agencies.
Evidence sourceRecommendations from scientific societies.35–43
JustificationThe goal of obesity treatment, as a chronic disease, is not only weight loss but also inducing favorable changes in body composition and improving complications and quality of life without clinically relevant side effects. Regulatory agencies certify that all approved obesity drugs meet safety and efficacy profiles based on established criteria. Due to past experiences with some approved obesity drugs that were later withdrawn for severe adverse effects, regulatory agencies now require adequate safety profiles, particularly cardiovascular, for long-term treatment.
Recommendation #2Do not prescribe or maintain pharmacological treatments for obesity during pregnancy and lactation.
Evidence sourceRecommendations from scientific societies.44–48
JustificationLifestyle interventions are recommended for pregnant and lactating women to achieve and maintain a healthy weight. Pharmacological treatments with approved drugs are not indicated during pregnancy, lactation, or for women planning pregnancy due to a lack of safety studies.
Recommendation #3Do not screen for hypercortisolism in all patients with obesity.
Evidence sourceRecommendations from scientific societies.49–61
JustificationCurrent clinical practice guidelines indicate that screening for hypercortisolism is not warranted in all obese patients without specific symptoms. Obesity alone has a very low pretest probability for Cushing's syndrome. Additionally, biochemical screening tests may yield false positives in obese patients.
NutritionRecommendation #1Do not use albumin levels to assess nutritional status in the acute postoperative period or in hospitalized patients.
Evidence sourceRecommendations from scientific societies.62,63
JustificationScientific evidence indicates that albumin levels do not reflect nutritional status but are influenced by factors such as inflammation, hydration, liver function, and protein synthesis. Albumin, a long half-life protein, does not change rapidly with nutritional intervention. Its use as a nutritional parameter in postoperative or hospitalized patients is neither appropriate nor efficient and may lead to overtreatment or undertreatment, increasing costs.
Recommendation #2Do not measure skinfold thickness in patients with bioimpedance analysis available.
Evidence sourceExpert consensus recommendations.37,64,65
JustificationBioimpedance analysis is a more precise technique for assessing body composition, providing detailed information on body fat, muscle mass, and water content. For evaluating and monitoring patients with malnutrition or obesity, bioimpedance offers more comprehensive and accurate data. Performing both techniques on the same patient adds no additional information and requires more time and resources.
Recommendation #3Do not prescribe pureed diets for hospitalized patients who can consume soft or regular-textured diets.
Evidence sourceRecommendations from scientific societies.66–68
JustificationESPEN guidelines emphasize meeting nutritional requirements with the least restrictions to ensure adequate intake. Pureed diets can cause dysphagia due to disuse atrophy of the masseter muscle and malnutrition due to limited intake. Their prescription should be limited to patients unable to consume other oral diets.
Pituitary diseaseRecommendation #1Avoid performing pituitary MRI in patients with hyperprolactinemia without first ruling out physiological causes, systemic diseases, drugs, or improper blood draw.
Evidence sourceGrading of Recommendations, Assessment, Development, and Evaluation (GRADE).69–79
JustificationHyperprolactinemia can result from analytical interference, physiological responses to pregnancy, lactation, or stress, or pathological causes such as drugs—antidepressants, antipsychotics, antiemetics, estrogens—or sellar tumors, including prolactinomas or other tumors with infundibular compression. Since 16% of the general population has clinically inert pituitary lesions, proper clinical and analytical evaluation is essential before requesting an MRI.
Recommendation #2Do not measure basal growth hormone (GH) levels in short stature (SS) evaluations.
Evidence sourceClinical guidelines using GRADE methodology.80–83
JustificationGH is a pulsatile hormone, and basal levels do not guide the diagnosis of GH deficiency. SS evaluation should include anthropometric parameters, clinical examination to rule out dysmorphic features, and psychomotor development assessment. If SS is confirmed, more informative tests include hand radiography, blood count, ESR, creatinine, electrolytes, calcium, phosphate, alkaline phosphatase, albumin, thyrotropin (TSH), free thyroxine, insulin-like growth factor 1 (IGF-1), anti-transglutaminase antibodies, and karyotype.
Recommendation #3Do not perform a stimulation test for adult GH deficiency diagnosis if decreased IGF-1 and at least 3 other pituitary hormone deficiencies are present.
Evidence sourceClinical practice guidelines using GRADE methodology.84–89
JustificationIn patients with 3 other pituitary hormone deficiencies (among corticotropin, TSH, gonadotropins, and/or antidiuretic hormone), there is a high pretest probability of adult GH deficiency. In this context, decreased IGF-1 levels confirm diagnosis, avoiding the need for a stimulation test. Consider potential false positives and negatives when interpreting IGF-1 levels.
Neuroendocrine tumorsRecommendation #1Do not request pancreatic imaging in individuals without diabetes mellitus, with hypoglycemia symptoms, in whom Whipple's triad or endogenous hyperinsulinism has not been confirmed as the cause.
Evidence sourceClinical guidelines using GRADE methodology.90
JustificationHypoglycemic syndromes are very rare in individuals who do not receive insulin or sulfonylureas as antidiabetic treatment. It is important to first confirm the hypoglycemic syndrome using the Whipple's triad (neuroglycopenic symptoms, coincidence of these symptoms with a laboratory glucose level <45 mg/dL, and symptomatic resolution upon treating the hypoglycemia), either spontaneously or during a fasting test of up to 72 h. It must also be confirmed that the hypoglycemia is due to endogenous hyperinsulinism (insulin and C-peptide not suppressed during hypoglycemia). Only if this triad with endogenous hyperinsulinism has been confirmed should a pancreatic MRI be requested to rule out an insulinoma.
Recommendation #2Do not measure chromogranin A (CgA) without discontinuing proton pump inhibitors (PPIs) in advance.
Evidence sourceClinical guidelines using GRADE methodology.91–98
JustificationCgA is a secretion protein from neuroendocrine cells and neurons that can be elevated in individuals with tumors or hyperplasia of various endocrine cell types, as well as in epithelial tumors. CgA has high sensitivity and low specificity for detecting neuroendocrine tumors. Treatment with PPIs leads to achlorhydria and, secondarily, a hyperplasia of enterocromaffin-like gastric cells, which increase CgA levels. Therefore, especially when using this marker for the initial diagnosis of neuroendocrine tumors, it is important to discontinue these drugs or replace them with anti-H2 drugs, at least, 2 weeks before determining this marker.
Thyroid diseaseRecommendation #1Do not use suppressive levothyroxine therapy in low-risk differentiated thyroid carcinoma patients with excellent response.
Evidence sourceRecommendations from scientific societies.99–106
JustificationRecent clinical practice guidelines classify TSH target levels based on initial recurrence risk and clinical response (dynamic risk stratification) after initial treatment. They aim to balance suppressive levothyroxine therapy with the risks of iatrogenic thyrotoxicosis. In patients with an excellent treatment response, serum TSH should be maintained in the low-normal range (0.3–2 mU/L).
Recommendation #2Avoid treating asymptomatic benign thyroid nodules with normal thyroid function.
Evidence sourceRecommendations from scientific societies.107–113
JustificationTreatment of benign thyroid nodules is only necessary for those with autonomous thyroid function, compressive symptoms, or esthetnic concerns. Clinical practice guidelines recommend clinical follow-up without treatment for cytologically benign, asymptomatic nodules, including those with mild size increase. Thyroid hormone therapy is not indicated in euthyroid patients with thyroid nodular disease.
Menopausal hormone therapyRecommendation #1Do not systematically deny hormonal and non-hormonal menopausal therapy to symptomatic women younger than 60 years with < 10 years since menopause onset.
Evidence sourceRecommendations from scientific societies and systematic reviews.114–132
JustificationThe positioning of scientific societies and professional organizations provides excellent resources for multidisciplinary teams (including gynecologists, primary care physicians, endocrinologists, nurses, and others) who care for middle-aged and older women. It is important for clinicians to recognize the early signs and symptoms of the menopausal transition and be prepared to offer treatment to mitigate these symptoms. Hormonal and non-hormonal therapies have been described, and it has been suggested that early initiation of hormonal therapy plays a substantial beneficial role in controlling symptoms, but also in improving quality of life and reducing mortality.
Recommendation #2Do not use pellets for menopause.
Evidence sourceRecommendations from scientific societies and systematic reviews.133,134
JustificationPellets are therapies not approved by the Food and Drug Administration (FDA), as they lack data on pharmacokinetics, safety, and efficacy, and do not have a label detailing the risks. Moreover, the therapy provides women with a supraphysiological hormonal dose, which raises safety concerns. To truly understand the benefits and risks of bioidentical hormonal therapy in the form of pellets for menopause, randomized trials with long-term follow-up are needed to compare bioidentical products with approved hormonal therapies for menopause.
ConclusionsSEEN 2023-2027 strategic reflection,135 under strategic line 3, aims to improve care and health outcomes for individuals with endocrine and/or nutritional diseases through quality, safety, and humanization. It promotes improving the quality of care in endocrinology and nutrition services by developing unit standards and performance indicators for comparative analysis (benchmarking) and emulation-based improvement.
The purpose of this document is to provide clear recommendations to endocrinology professionals that have a direct and immediate application in their clinical practice. The recommendations have been selected and drafted by members designated by the various WGs and KAs of the Spanish Society of Endocrinology and Nutrition (SEEN) who participated in the document. Although, obviously, it is not possible to cover the full scope of our specialty, the topics addressed and the recommendations formulated are of clear use for the everyday practice of clinical endocrinologists.
All the recommendations stated here are supported by previously published documents, such as clinical practice guidelines, systematic reviews, and randomized clinical trials. Recommendations based on indirect evidence—ie, studies in which patients differ from those for whom the recommendation is intended, or those where the procedure is not specifically related to the recommendation—have been avoided.
The main limitation of this document comes from the lack of recommendations in some areas of the specialty. Although a recommendation for pediatric patients has been included, most of the recommendations presented are for adult patients. A common methodology for searching the bibliographic information and selecting documents for the drafting of recommendations by the various participating areas and groups has not been used. This document allowed the selected experts to use their clinical judgment to collect the thematic areas they considered most appropriate and useful to improve clinical practice in endocrinology. In addition, a rigorous review process was followed, both for the selected recommendations and for the justificatory texts. Two reviews were conducted by the SEEN groups and areas, and finally, a review by the SEEN board of directors, which approved the document.
Despite the importance of various awareness campaigns to prevent inappropriate practices, knowledge and implementation of these in the routine clinical practice is rather limited.8 Some studies have even shown only a moderate impact from the implementation of the proposed recommendations.136 The authors of this document believe that its publication and dissemination through a scientific society like SEEN, with more than 2,500 members, most of whom provide health care as their main professional activity, is an appropriate procedure to reach a large number of endocrinologists who can apply the recommendations to their routine clinical practice in Spain. Therefore, this document reflects the result of joint and coordinated work from various SEEN KAs and WGs, formulating different recommendations for clinical practice in the field of endocrinology and nutrition.
Ethical responsibilitiesAs a literature review and WG coordination study, ethical committee approval was deemed unnecesary. However, the project was approved by the SEEN board before drafting.
The full document is available at: https://manual.seen.es/article?id=6763e8fe-aed0-43a3-a6ae-767b0aca0133137
FundingNone declared.
Authors wish to thank the coordinators of SEEN KA (diabetes, obesity, nutrition, neuroendocrinology, and thyroid) and the gonads, identity, and sexual differentiation working group for their collaboration in selecting authors and reviewing recommendation texts, and the board of directors of SEEN for approving the project and reviewing the manuscript.
President: Ignacio Bernabéu; First Vice President: Juan José Díez; Second Vice President: Gabriel Olveira; Treasurer: María Alba Galdón Sanz-Pastor; Secretary: Alberto Fernández Martínez; Members: Enma Anda Apiñaniz, Manuel Gahete Ortiz, Nuria Vilarrasa García, Julia Ocón Bretón, Juan José López Gómez, Beatriz Lardiés Sánchez, and Francisco Pita Gutiérrez.
