Investigating the influence of adherence to dietary behaviour recommendations on percentage total weight loss after metabolic and bariatric surgery: study protocol for the OBSERVE study, a single-centre observational, prospective, longitudinal cohort study in Germany

Abstract

Introduction

The long-term success of metabolic and bariatric surgery (MBS) depends largely on adherence to health-promoting behaviour following MBS. Especially, adherence to healthy dietary behaviour in line with guidelines appears to be of the utmost importance. The primary objective of the present study is to evaluate the effect of adherence to dietary behaviour recommendations and the percentage of total weight loss (%TWL) after MBS. Adherence is hypothesised to have a positive influence on the %TWL, 24 months after MBS. Furthermore, we investigate the association of various sociodemographic, organisational, psychological and behavioural factors prior to and after MBS and their effect on %TWL.

Methods and analysis

The present study is a single-centre observational, prospective, longitudinal cohort study conducted in Germany. Data are collected at nine measurement points (T0: 4 weeks prior to MBS; T1: 2 weeks after MBS, T2: 5 weeks after MBS; T3: 3 months after MBS; T4: 6 months after MBS; T5: 12 months after MBS; T6: 18 months after MBS; T7: 24 months after MBS; and T8: 36 months after MBS). Adherence to dietary behaviour recommendations is assessed using the Dietary Behavior Inventory-Surgery (DBI-S). N=325 patients applying for MBS will be included in the study. A regression analysis approach is chosen to answer the primary research question. The primary outcome %TWL is regressed at T7 (24 months after MBS) in a causal analysis on dietary adherence (DBI-S score) at T3–T7, with the covariates age, gender, marital status, educational attainment, employment status, Patient Health Questionnaire-4 score and body mass index at T0 and MBS method at T1. Stepwise hierarchical regression analyses are performed and analysed for significant model differences using χ² difference tests. Effect sizes are estimated by R². Group differences are analysed using t-tests and Analyses of variance (ANOVAs). Bivariate correlations of continuous variables are examined using regression/correlation analyses.

Ethics and dissemination

The Ethics Committee of the Medical Faculty of the University of Essen-Duisburg has approved the conduct of the study (24-11969-BO). Results will be disseminated through manuscripts in clinical/academic peer-reviewed journals, presentations at academic conferences and communications with partners, participants and other stakeholders. Key findings will also be published in lay language on a publicly accessible website and disseminated via various (social) media channels.

Trial registration number

The study has been prospectively registered on 8 October 2024 in the German Clinical Trials Register (DRKS00034888).

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• Observational, longitudinal study design to assess relevant aspects over a 3-year timespan.

• Use of validated measurement instruments.

• Possible high drop-out rate due to long follow-ups.

• The use of medication for obesity is not included as a factor in this study.

• Patients’ weight will only be measured at initiation and assessed on the basis of self-reported data at the other measurement points.

Introduction

In 2022, one in eight people worldwide was living with obesity, defined as a body mass index (BMI) greater than or equal to 30 kg/m².¹ The WHO defines obesity as ‘abnormal or excessive fat accumulation that poses a risk to health’.² The World Obesity Federation has declared obesity to be a chronic, relapsing and progressive disease.³ The prevalence of obesity has been growing rapidly for years. Obesity has reached epidemic proportions worldwide, accounting for approximately 650 million adults with obesity.⁴ There is greater morbidity and mortality in patients with severe obesity, especially classes II (BMI 35.0 kg/m² to 39.9 kg/m²) and III (BMI ≥40 kg/m²). Obesity is a recognised risk factor for the development of somatic and psychological comorbid conditions such as cardiovascular disease, type 2 diabetes mellitus, malignancy, asthma, osteoarthritis, chronic back pain, obstructive sleep apnoea, non-alcoholic fatty liver disease, gallbladder diseases, depression, anxiety disorders and eating disorders.^{5 6}

Using solely conservative treatment methods (eg, dietary counselling, increasing physical activity, behavioural adaptations), only a small proportion of patients with obesity are enabled to reduce their weight permanently. Previous studies showed that, on average, only 1 male in 210 with grade I obesity succeeds in achieving a normal weight. For females, the ratio is 1 in 124 individuals.⁷ The probability of a person with obesity achieving normal weight through conservative weight loss methods decreases further with increasing BMI.⁸

Metabolic and bariatric surgery (MBS) is one of the most effective treatment methods for severe obesity in terms of long-term weight reduction,⁹ reduction of comorbid diseases and mortality.¹⁰ For this reason, the current European guidelines for obesity in adults recommend MBS for patients with a sufficiently high BMI (≥35 kg/m²) and existing obesity-associated comorbidities as well as for patients with a BMI ≥40 kg/m² without obesity-related comorbidities.¹¹ Recent guidelines developed by the world’s leading authorities (The International Federation for the Surgery of Obesity and Metabolic Disorders and The American Society for Metabolic and Bariatric Surgery) on MBS even recommend MBS for individuals with a BMI >35 kg/m², regardless of presence, absence or severity of comorbidities.¹² Nevertheless, many patients experience long-term complications, such as only insufficient weight loss (defined as an excess weight reduction of less than 50%),¹³ weight regain (defined as an increase of more than 25% from nadir weight) or negative psychological effects such as depression, anxiety or eating disorders.¹⁴

Current evidence suggests that especially the patients’ adherence to dietary behaviour recommendations (encompassing the categories ‘food choice, eating behaviour, and dietary intake/nutrition’) is an indicator for long-term success of MBS with regard to weight loss and health outcomes.¹⁵ The Dietary Behavior Inventory-Surgery (DBI-S) is an instrument to measure adherence to dietary recommendations after MBS and was constructed based on nutrition recommendations for patients after bariatric surgery aggregated by Sherf-Dagan and colleagues.¹⁶ The DBI-S can be used to assess adherence to dietary behaviour recommendations in an efficient and valid way.

In a cross-sectional study conducted among 485 individuals after MBS,¹⁷ a positive association was found between adherence to dietary behaviour recommendations, as measured by the DBI-S, and percentage total weight loss (%TWL) after MBS. Furthermore, it was shown that adherence to dietary behaviour recommendations moderates the effect between time since MBS and %TWL with several covariates identified in the study. Based on these results, the proposed longitudinal OBSERVE study aims to explore the causal relationship between adherence to dietary behaviour recommendations after MBS (measured with the DBI-S) and the %TWL.

According to empirical data, the long-term success of MBS also depends on psychological factors such as self-efficacy, distress, quality of life, impulsivity, physical/mental health and eating behaviour in addition to the physiological adaptations caused by the surgery.18,20 These factors will also be investigated in the OBSERVE study in relation to their influence on %TWL after MBS and its progression over time.

A study analysing organisational factors (eg, the quality of the hospital discharge process) has shown that a high-quality discharge process is associated with better health outcomes and fewer unplanned hospital re-admissions after MBS.²¹ Therefore, the quality of the hospital discharge process will be additionally investigated in the OBSERVE study.

Objectives

The primary objective of the study is to evaluate the influence of adherence to dietary behaviour recommendations and guidelines (measured with the DBI-S) on %TWL after MBS. Adherence is hypothesised to have a positive influence on the %TWL, 24 months after MBS (T7). Weight regain usually occurs 18–24 months after MBS.²² This makes it adequate to use T7 (24 months after MBS) for assessing weight regain validly in the proposed study.

The second aim is to investigate the association of various sociodemographic and medical (age, gender, BMI prior to MBS, highest weight, lowest weight, marital status, formal education, MBS procedure), psychological (self-efficacy, depression and anxiety symptoms, body image, food cravings) and behavioural factors (participation in aftercare programmes and self-help groups) prior to and after MBS and their effect on %TWL and other health outcomes, as well as relevant mediators and moderators of these relationships.

Further study goals are the investigation of the association between the quality of the discharge process and complications occurring after surgery (patient safety indicators, for example, necessary re-interventions, unplanned hospital admissions, rebound), as well as health-related and nutrition-related outcomes (nutrition knowledge and eHealth literacy) and %TWL after MBS.

Materials and methods

As to our knowledge, no specific guideline for the reporting of study protocols describing longitudinal observational cohort studies exists.²³ Due to the non-interventional nature of the planned study, this study protocol is reported according to the ‘The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement’.²³ The study has been prospectively registered in the German Clinical Trials Register (DRKS00034888) on 8 October 2024. In case of important modifications regarding the trial design, the ethics committee will be informed and the trial registration will be updated.

Study design and procedure

The present study is a single-centre observational, prospective, longitudinal cohort study conducted in Germany to assess patients planning to undergo before and after MBS procedure. Patients will be assessed regarding different parameters (eg, weight, adherence, psychological variables) at fixed intervals (see table 1 for details). All data are collected via self-reports from the patients via digital case report forms.

Table 1. Assessment schedule of the OBSERVE study.

Measures	Planned assessment time points
	T0	T1	T2	T3	T4	T5	T6	T7	T8
Demographic and medical characteristics
Age	X
Height	X
Lowest weight	X
Highest weight	X
Formal education	X
Weight	X	X	X	X	X	X	X	X	X
Employment	X	X	X	X	X	X	X	X	X
Marital status	X	X	X	X	X	X	X	X	X
MBS procedure	X	X
Primary outcome
%TWL/%EWL				X	X	X	X	X	X
Secondary outcomes
Nadir weight				X	X	X	X	X	X
Quality of life	X	X	X	X	X	X	X	X	X
eHealth literacy	X	X	X	X	X	X	X	X	X
General nutritional knowledge	X	X	X	X	X	X	X	X	X
Body image	X	X	X	X	X	X	X	X	X
Dietary behaviour				X	X	X	X	X	X
Food cravings	X	X	X	X	X	X	X	X	X
Care transitions		X
Patient safety/complications			X	X	X	X	X	X	X
Self-efficacy	X	X	X	X	X	X	X	X	X
Patient health	X	X	X	X	X	X	X	X	X
Distress	X	X	X	X	X	X	X	X	X
Participation in aftercare programme		X	X	X	X	X	X	X	X
Participation in self-help group	X	X	X	X	X	X	X	X	X
Body contouring procedures	X	X	X	X	X	X	X	X	X

T0=4 weeks prior to MBS; T1=2 weeks after MBS; T2=5 weeks after MBS; T3=3 months after MBS; T4=6 months after MBS; T5=12 months after MBS; T6=18 months after MBS; T7=24 months after MBS; T8=36 months after MBS; Quality of Life=Bariatric Quality of Life questionnaire (BQL); eHealth Literacy=eHealth Literacy Scale (GR-eHEALS); General Nutritional Knowledge=General Nutrition Knowledge Questionnaire (GNK-Q); Body image=Body-Questionnaire (Body-Q); Dietary Behaviour=Dietary Behavior Inventory-Surgery (DBI-S); Food Cravings=Short version of Food Cravings Questionnaire (FCQ-T); Care Transitions=Care Transitions Measure (CTM); Self-efficacy=General Self-Efficacy Short Scale-3 (GSE-3); Patient Health=Patient Health Questionnaire-4 (PHQ-4).

The proposed study incorporates nine distinct measurement time points: the baseline assessment 4 weeks prior to MBS (T0), a post-MBS assessment 2 weeks after MBS (T1) and an assessment 5 weeks after MBS (T2). Furthermore, assessments will take place at 3 (T3), 6 (T4), 12 (T5), 18 (T6), 24 (T7) and 36 months (T8) after MBS. Participants receive an e-mail with a personalised link to the digital case report forms at each measurement time point. Data will be collected and stored via the institute’s own data management and patient-reported outcome measurement system (‘Promspace’). To ensure adherence and fidelity, all participants will receive notifications that inform them when a new assessment is due, including a link to access the assessment. Patients receive additional reminders if they have not completed the assessment 1 week after it has been due in order to enhance motivation to log in. All notifications and reminders are standardised and sent via email. See figure 1 for the schedule of the study.

Figure 1. Study flow chart of the OBSERVE study. MBS, metabolic and bariatric surgery.

The participants complete different sets of questionnaires at each of the different measurement points. Some values are collected once, others at multiple measurement points. See table 1 for the full assessment schedule of the OBSERVE study.

Participant eligibility and recruitment

Patients will be included if they are at least 18 years old, have a BMI ≥35 kg/m², have access to an Internet-capable device and are in the process of seeking MBS. Furthermore, they have to provide written informed consent after being informed about the procedure and goal of the study.

Patients will be excluded if they do not have sufficient knowledge of the German language. Patients will also be excluded if they have already undergone MBS in the past. No further exclusion criteria will be applied.

Patients will be recruited via personal contact during an appointment for a psychological evaluation before the planned MBS at the Clinic for Psychosomatic Medicine and Psychotherapy, LVR-University Hospital Essen, University of Duisburg-Essen. Patients who meet the inclusion criteria will be informed about the possibility of participating in the OBSERVE study. Flyers and study information will be handed out by clinical providers. Patients will have the opportunity to ask questions about the study and will be informed about their right to withdraw from participation at any time. They receive sufficient information and time to decide whether to take part in the study or not. After patients have provided written informed consent, the inclusion of the patients into the study is conducted by the study team.

Outcome measures

The primary outcome measure is the percentage total weight loss (%TWL) = [(pre-op weight − post-op weight) / (pre-op weight)] × 100, 24 months after MBS (T7). This primary endpoint was chosen because weight loss is usually complete and stabilised by this time.¹³

Secondary outcome measures are other weight-related outcomes (nadir weight, %EWL), health-related outcomes (quality of life, body image, self-efficacy, depression and anxiety symptoms), patient safety indicators, quality of the discharge process after MBS, and eHealth literacy. In addition to the analysis of the impact of dietary adherence after MBS on these factors, secondary outcome measures are integrated into statistical analyses as covariates.

Description of the measurement instruments

Sociodemographic and medical data

The sociodemographic data of sex, age, formal education/vocational training as well as medical data (height, lowest weight at current height, highest weight at current height) are routinely recorded at the initial consultation in our outpatient clinic. Employment and marital status are recorded at every measurement point to capture possible changes.

Weight pre/post/progression

Single items are used to record the progression of weight at all measurement points. This information can be used to calculate the BMI at the relevant time. Baseline weight is measured with a Seca bio impedance analysis scale.

Dietary adherence

The DBI-S is used to assess dietary adherence.¹⁶ The DBI-S comprises 13 items and is a validated instrument for recording dietary adherence after MBS in accordance with international medical recommendations and the European Guidelines on metabolic and bariatric surgery.¹¹

Quality of life

Quality of life is assessed using the validated Metabolic and Bariatric Diseases - Quality of Life Questionnaire.²⁴ It consists of 30 items. Quality of life is assessed at all measurement points.

eHealth literacy

The validated German version of the eHealth Literacy Scale is used to record eHealth literacy.²⁵ The scale consists of eight items.

Nutritional knowledge

Nutritional knowledge is assessed using the validated Nutrition Knowledge Questionnaire.²⁶ It measures general nutritional knowledge using 20 items on a three-point scale with answer options being true/false/don’t know.

Body image

Body image is assessed using the validated ‘Body image’ subscale of the Body-Q.²⁷ This seven-item scale measures body image and is designed for patients after MBS. Patients have to indicate for each statement whether they agree or do not agree with it.

Impulsive eating behaviour

Impulsive eating behaviour is assessed using the short version of the Food Cravings Questionnaire-Trait (FCQ-T).²⁸ The FCQ-T is a 39-item questionnaire that assesses food cravings as a personality trait. The items are rated on a 6-point Likert scale.

Quality of the discharge process

The quality of the discharge process is measured using the Care Transitions Measure.²⁹ The instrument is the most frequently used worldwide to measure the quality of the discharge process from the patient’s perspective. With 15 items, four dimensions of patient-centredness in discharge management are recorded. A 5-point Likert scale is used (from 1 = ‘Do not agree at all’ to 4 = ‘Completely agree’ and 5 = ‘I don’t know/I don’t remember/doesn’t apply’).

Patient safety indicators

Patient safety indicators are used to record whether patients have experienced adverse/unplanned events after surgery and discharge from hospital. As there are no validated instruments, single self-generated items are used. The items have already been used in previous studies that are not yet published. In addition to recording whether unplanned hospital readmissions were necessary or medication complications occurred, typical adverse events that can occur after bariatric surgery were also assessed (eg, suture insufficiency, nausea/vomiting, cardiological problems (heart attack, cardiac arrhythmia)).

Self-efficacy

The General Self-Efficacy Short Scale-3 (GSE-3) is used to assess perceived self-efficacy.³⁰ The GSE-3 consists of three statements and is a validated and efficient instrument for the assessment of self-efficacy. Items are answered on a 5-point rating scale, comprising the categories 1=does not apply at all, 2=applies a bit, 3=applies somewhat, 4=applies mostly and 5=applies completely.

Depression and anxiety symptoms

The Patient Health Questionnaire-4 (PHQ-4) is used to assess depression and anxiety symptoms.³¹ The PHQ-4 consists of four items and is a validated instrument for the dimensional assessment of depression severity. The PHQ-4 (Patient Health Questionnaire with four questions) asks about the frequency with which respondents have felt impaired by certain complaints in the past two weeks. The PHQ-4 offers four response options per question, each with a numerical value from 0 to 3. The answers are as follows: ‘not at all’ (0), ‘some days’ (1), ‘more than half of the days’ (2) and ‘almost every day’ (3).

Resilience

Distress is assessed by using the Distress Thermometer.³² It consists of one item and a scale from 0 to 10 to assess the severity of distress symptoms an individual experiences. Additionally, the respondent is prompted to identify sources of distress using the provided problem list.

Participation in pre-examination and follow-up examination

Participation in pre-examination and follow-up examination is recorded using self-developed single items. In addition, the questionnaire assesses which specialist carries out the follow-up care.

Participation in self-help group

Participation in self-help group is assessed using a self-developed single item (‘Are you participating in a self-help group at the moment?’).

Reconstructive surgery

The intention to undergo body contouring surgery (reconstructive surgery) is recorded using a self-developed item. If reconstructive surgery has already taken place, the type and date of this are asked.

Sample size calculation

The sample size calculation is based on the statistical examination of the primary outcome (%TWL, 24 months after MBS). For the prediction of the primary outcome, adherence to dietary guidelines will be entered as the main predictor. Furthermore, the covariates age, gender, marital status, educational attainment, employment status, lowest and highest weight, height, PHQ-4 score, BMI (at T0) and MBS method (at T1) are to be included. In summary, a total of 13 predictors are included in a hierarchical multiple regression model. The following statistical assumptions are made: significance level of α = 0.05; test power of 1-β = 0.90 (the probability of the overall model to find an effect that actually exists); an effect size of the overall model of R² = 0.30 derived from previous research.¹² These assumptions result in a necessary total sample size of N=65. We expect the sample loss up to the measurement end time point (T7, 24 months after MBS) of the primary outcome to be 80%. Taking this sample attrition into account, a final sample size of N=325 (n=65/(100%–80% attrition)) at T0 was determined.

Statistical methods

Descriptive methods are used to present distribution and frequencies (for categorical variables) as well as mean values and standard deviations (for continuous variables). Group differences (eg, between gender, marital status or educational attainment) are analysed using t-tests and Analyses of variance (ANOVAs). Bonferroni corrections for multiple testing will be applied. Bivariate correlations of continuous variables are examined using regression/correlation analyses.

A hierarchical multiple regression analysis approach is chosen to answer the primary research question. In addition to the predictor (adherence to dietary guidelines), relevant variables are considered as covariates. The primary outcome (%TWL) at T7 (24 months after surgery) is regressed on dietary adherence at T3–T7 and the covariates age, gender, marital status, educational attainment, employment status, PHQ-4 score, BMI at (T0) and MBS method at (T1). Stepwise hierarchical regression analyses are performed and analysed for significant model differences using χ² difference tests. Effect sizes of the overall models are estimated by R². Unstandardised (b) and standardised (β) regression coefficients will be used to evaluate the effect of each predictor.

The secondary research questions will be investigated using hierarchical multiple regression analyses approaches. To gain a deeper understanding of changes over time after MBS of all study predictors and outcomes, time series analyses will be performed. These analyses are conducted to identify whether predictors and outcomes follow linear trends or non-linear trends (eg, sudden gains and losses) or are random. This approach helps to understand the long-term outcomes and determinants of patients after MBS.

Patient and public involvement

Neither patients nor the public were directly involved in the development of this study design. However, preliminary results of a quantitative study currently conducted at our institute examining the needs and expectations of patients after MBS regarding medical care indicate that patients see the need for a greater focus on physical and especially mental health after MBS. The findings of the planned longitudinal OBSERVE study will contribute to the fulfilment of the patients’ needs. Furthermore, a preliminary cross-sectional study including the target patient population delivered the foundation of the study.¹⁵

Ethics and dissemination

The Ethics Committee of the Medical Faculty of the University of Essen-Duisburg has approved the conduct of the study (24-11969-BO). The study will be conducted in accordance with the WHO Handbook on good clinical research practice.³³

Written informed consent is obligatory for all patients to participate in the OBSERVE study and is obtained after providing oral and written information on the study. All patients will be able to withdraw from participation at any time without any disadvantage. No intervention will be conducted during the study. Solely, the role of MBS as a predictor for the above-mentioned factors and their correlations will be observed. However, the conduct of the MBS is not part of the present study. Therefore, the study poses no distinct ethical concerns from our point of view, as approved by the Ethics Committee. We are unaware of specific risks or disadvantages that might affect patients during the study, and we expect no specific adverse or serious adverse events resulting from participating in the OBSERVE study. The study will be performed in accordance with the recommendations guiding ethical research involving human subjects adopted by the 18th World Medical Association General Assembly, Helsinki, Finland, June 1964, amended at the 64th World Medical Association General Assembly, Fortaleza, Brazil, 2013.

Results will be disseminated through manuscripts in clinical and academic peer-reviewed journals, presentations at academic conferences and communications with partners, participants and other stakeholders. Key findings will also be published in lay language on a publicly accessible website and disseminated via various (social) media channels.

Discussion

Obesity is a global health problem of pandemic proportions. The proposed longitudinal OBSERVE study will investigate the extent to which adherence to dietary adherence after MBS influences %TWL and which factors moderate this relationship in the long term. The OBSERVE study will also investigate how psychological well-being develops up to three years after the operation and which pre-operative and post-operative factors influence this.

The findings will then be used to identify factors that affect weight loss as well as mental and physical well-being after MBS and to develop perioperative interventions to positively influence health and well-being. As obesity is a multifactorial complex disease, a holistic approach is necessary to identify all factors and predictors that are relevant for the success of MBS in terms of long-term weight loss and well-being of the patients.

A particular strength of the study is the longitudinal study design, which makes it possible to examine the relevant factors pre-MBS and post-MBS over a period of more than three years, as well as their dynamic changes. The sample is comprised of patients who were treated in different clinics using different MBS approaches.

An inherent problem in longitudinal studies is the potentially high drop-out rate over the course of the study. In order to keep this rate as low as possible, participants are sent individual links at each measure point and reminded to participate. Participants also have the opportunity to contact the study team throughout the entire duration of the study to ask questions or report any difficulties in connection with the study participation.

Furthermore, all data generated is self-reported. It must be taken into account that information, especially weight information, is not always given correctly by participants. However, the participants have a particular focus on their body weight, so it can be assumed that this information is correctly observed and reported.

Although dietary counselling is not standardised prior to and after MBS, dietary adherence according to MBS is a central point of the study. By recording dietary knowledge, it can be ensured that a fundamental lack of knowledge about dietary guidelines can be taken into account as a possible confounder and corrected for this.

In the current study, the use of obesity medication (before or after MBS) such as semaglutide or tirzepatide is not being assessed. However, as an increasingly large group of patients use this medication before and after MBS, future studies should include and investigate their effect.

In summary, the study will contribute to the understanding of long-term psychological and behavioural factors in the pre-operative and post-operative context in patients undergoing MBS. Especially, the role of adherence to dietary behavior recommendations will be investigated to ensure adequate nutritional advice after MBS, thereby providing an important foundation for the evidence-based improvement of existing and new health services for individuals with obesity.