Well-being and dietary adherence in patients with coeliac disease depending on follow-up

(1)

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=igas20

Scandinavian Journal of Gastroenterology

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/igas20

Well-being and dietary adherence in patients with

coeliac disease depending on follow-up

Jesper Lexner, Henrik Hjortswang, Rickard Ekesbo & Klas Sjöberg

To cite this article: Jesper Lexner, Henrik Hjortswang, Rickard Ekesbo & Klas Sjöberg (2021) Well-being and dietary adherence in patients with coeliac disease depending on follow-up, Scandinavian Journal of Gastroenterology, 56:4, 382-390, DOI: 10.1080/00365521.2021.1889024

To link to this article: https://doi.org/10.1080/00365521.2021.1889024

Published online: 23 Feb 2021.

Submit your article to this journal

Article views: 316

View related articles

(2)

ORIGINAL ARTICLE

Well-being and dietary adherence in patients with coeliac disease depending on

follow-up

Jesper Lexnera,b, Henrik Hjortswangc,d, Rickard Ekesboeand Klas Sj€oberga,b a

Department of Clinical Sciences, Lund University, Lund, Sweden;bDepartment of Gastroenterology, Skåne University Hospital, Malm€o, Sweden;cDepartment of Gastroenterology and Hepatology, Link€oping University, Link€oping, Sweden;dDepartment of Health, Medicine, and Caring Sciences, Link€oping University, Link€oping, Sweden;eDepartment of Clinical Sciences, Lund University, Vårdhuset Malm€o, Sweden

ABSTRACT

Objective: It is not clear how follow-up of coeliac disease should be optimally organised. In Malm€o, Sweden, patients are followed up by general practitioners (GP), but in Link€oping by gastroenterolo-gists (GE). The aim of this study was to investigate if there were any differences in well-being and diet-ary adherence depending on type of follow-up.

Methods: All adult patients with newly diagnosed biopsy-verified coeliac disease in the cities between 2010 and 2014 were offered to participate. Data was retrieved comprising demography, laboratory analyses, questionnaires (Gastrointestinal Symptoms Rating Scale, Short Health Scale, Multidimensional Fatigue Inventory, Psychological General Well-being Index and Short Form 36) and follow-up.

Results: In the GP cohort 39/73 patients and in the GE cohort 58/121 agreed to participate (mean age 43 and 44 years, 69 and 60% women, respectively). A follow-up to a dietician was carried out in 31% and 93% of patients, respectively (p < .001). In the GP group 28% had eaten gluten-containing food during the last 4 weeks compared to 9% in the GE group (p ¼ .01). Despite this, no differences could be seen in vitamin or mineral levels. The questionnaires did not indicate any major discrepancies in subjective health.

Conclusion: Irrespective of the design of the follow-up physical and mental well-being were compar-able. Dietary adherence was not quite as good in the GP group but follow-up in a primary care setting can still be a suitable and equivalent alternative. However, it is crucial that the dietary counselling is structured in a way that ensures dietary adherence.

ARTICLE HISTORY

Received 2 September 2020 Revised 26 January 2021 Accepted 7 February 2021

KEYWORDS

Coeliac disease; diet; follow-up; quality of life; well-being

Introduction

Among autoimmune diseases, the prevalence of coeliac dis-ease (gluten intolerance) has incrdis-eased substantially over the last decades [1,2]. The worldwide prevalence is estimated to about 1%, and it affects women more often than men [3]. Coeliac disease is even more common in Sweden where around 2% might be affected [4]. Though coeliac disease traditionally has been considered a paediatric disease, the age of onset varies and symptoms may very well present in adulthood [5].

If undiagnosed or not properly treated coeliac disease may cause both physical and psychological complications. Some of those are related to malabsorption, such as iron or vitamin deficiencies [6]. As for psychiatric comorbidities, qual-ity of life can be impaired and there is an increased risk of depression [7,8]. Particularly women suffer from a reduced level of well-being, and those with more gastrointestinal symptoms usually have lower levels of well-being [9]. Quality of life, and maybe also depressive symptoms, tend to improve when a strict gluten-free diet is initiated [7,8].

Fatigue is another common extraintestinal symptom [10]. Due to its autoimmune nature and relation to HLA-DQ2 and -DQ8, patients with coeliac disease have an increased risk of other autoimmune diseases, such as type 1 diabetes mellitus and autoimmune thyroid diseases [11]. These examples illus-trate the importance of not just an early diagnosis, but also a qualified and thorough follow-up care afterwards both to ensure compliance to the gluten-free diet and to discover complications.

There is a lack of evidence on how the follow-up care of coeliac disease should be managed. This is especially true for those who present with coeliac disease as adults. As there is too little evidence on how to follow up the patients after the diagnosis and initiation of a gluten-free diet, practices have differed greatly between different healthcare providers [12]. There is a disagreement in international guidelines whether dietician consultations, determination of tissue transglutami-nase (IgA-tTg) antibody serology and basic laboratory blood tests to discover complications should be carried out annu-ally or only when considered necessary [13–16]. Neither is it clear whether the follow-up care should be provided by a

CONTACTKlas Sj€oberg klas.sjoberg@med.lu.se

ß 2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.

This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.

2021, VOL. 56, NO. 4, 382_–390

(3)

specialist in gastroenterology at the hospital or if it is suffi-cient to be followed up by a physician at a primary health-care centre.

Due to different economic resources and the lack of evi-dence-based guidelines on the follow-up care of coeliac dis-ease patients, practices differ between different healthcare providers in Sweden. In Malm€o the diagnostic process and follow-up care is managed by general practitioners (GP) at primary healthcare centres. In contrast, all patients in Link€oping are followed up by a specialist in gastroenterology (GE) at the university hospital until healing of the small intes-tine (regression of villous atrophy) has been confirmed by a small intestinal biopsy. In the Malm€o region control biopsies are rarely taken. The patients in Link€oping are also followed up by a dietician for 5 years and more frequently in the case of persisting villous atrophy. As of today, it is not known if any of these two practices result in a better outcome for the patients. The aim of this study was to investigate if there were any differences in physical and mental well-being and dietary adherence after 5–10 years follow-up in two cohorts with coeliac disease, depending on whether the follow-up care was provided at a primary healthcare centre or at a spe-cialised out-patient ward.

Patients and methods

In this Swedish multicentre study adult patients diagnosed with coeliac disease in Malm€o (GP led follow-up) and Link€oping (GE led follow-up) from 1 January 2010 to 31 December 2014 were invited to participate. Patients in the GP cohort were identified from registries of coeliac disease patients at the Department of Pathology at Skåne University Hospital, Malm€o. In Link€oping all patients are diagnosed by a specialist in gastroenterology, why patients in the GE cohort were identified from registers of coeliac disease-related visits at the department of gastroenterology. All patients had to have their diagnosis confirmed by biopsy and clinical evalu-ation during the aforementioned dates and had to be 18 years or older when the biopsy arrived at the department of pathology. The national recommendations for diagnosis at the time required verification with biopsies from the duode-num [17]. Both cities were selected because the health care at both sites was based on no more than one hospital, one department of pathology and specified general practitioners. That means that all patients diagnosed in the specific areas will be identified through the registers and invited regardless of how the follow-up is organized. The patients’ medical records were used to exclude patients who were diagnosed outside the time span or who were followed up outside the two chosen regions. Patients were also excluded if they were deceased or no longer lived in Sweden. A total of 73 patient-sin the GP cohort and 121 in the GE cohort met the inclusion criteria and were offered to participate in the study.

We investigated the health of the patients 5–10 years after their coeliac disease diagnosis. Analyses performed since 1 January 2019 were included, and if several analyses of the same type of test had been performed the latest available test was used. Data about demography, follow-up and

comorbidities was retrieved from the medical records and if needed from the patients themselves. The patients’ medical records were also used to retrieve results of laboratory analy-ses for IgA-tTg serology, vitamin and mineral deficiencies. Presence of gastrointestinal symptoms, general well-being and dietary adherence were evaluated by questionnaires (sent by mail).

IgA-tTg titres were analysed with enzyme-linked immuno-assay with human recombinant tTg-antigen according to established procedures (Thermo Fisher ELiACelikey IgA Well, Phadia Laboratory Systems). Levels of IgA-tTg < 7 kilo-arbi-trary units per litre (kU/L) were considered negative. IgA defi-ciency was excluded.

We used the Gastrointestinal Symptoms Rating Scale (GSRS) questionnaire to evaluate gastrointestinal symptoms the last week. The GSRS was initially developed for irritable bowel syndrome (IBS) and peptic ulcer disease as a 15-item questionnaire (GSRS-15) [18]. The 15 items can be grouped into five dimensions: abdominal pain, reflux, indigestion, diarrhoea and constipation. The questionnaire also exists in a 13-item version (GSRS-IBS) specific for IBS [19]. These items can be grouped into different dimensions; pain, bloating, constipation, diarrhoea and satiety [20]. As some of the items in the two questionnaires are the same, we used a 20-item questionnaire including all items from the two question-naires. Each item is graded on a 7-point Likert scale. All dimensions are presented as the sum of the included items divided by the number of questions. The results of the GSRS questionnaires can also be presented as the total sum of the items in the GSRS-15 and GSRS-IBS, respectively. The higher the total score is, the more severe are the gastrointes-tinal symptoms.

The Short Health Scale (SHS) is a four-item questionnaire. Initially developed for inflammatory bowel disease, the scale evaluates four subjective health dimensions related to the patient’s intestinal disease: gastrointestinal symptom burden, social function, disease-related worry and general well-being. Answers are graded on a 6-point Likert scale. The higher the score is, the more troubled is the patient by their gastro-intestinal disease [21].

The Multidimensional Fatigue Inventory (MFI) was used to assess the study participants’ experience of fatigue. The MFI is a 20-item self-report questionnaire. The results are presented in five subgroups: general fatigue, physical and mental fatigue, reduced motivation and reduced activity. A 5-point Likert scale is used, and a high score represent a high degree of fatigue [22].

The Psychological General Well-being Index (PGWB) was used to measure subjective well-being the last week. The PGWB includes 22 questions and answers are graded on a 6-point Likert scale. The 22 questions can be grouped into six sub-scales: anxiety, depressed mood, well-being, self-control, general health and vitality. A total sum of all the 22 ques-tions can also be calculated. Higher scores represent better well-being [23].

The Short Form 36 (SF-36) was used to evaluate the par-ticipants’ perception of their own health. SF-36 is a 36-item questionnaire which consists of eight sub-dimensions:

(4)

physical functioning, role limitations due to physical prob-lems, bodily pain, general health perceptions, vitality, social functioning, role limitations due to emotional problems and mental health. The answers were transformed according to a standard algorithm into a scale from 0 (worst possible health state) to 100 (best possible health state) [24]. The results were also presented as two summary scores, namely the Physical Component Summary score (PCS) and the Mental Component Summary score (MCS), which are standardised to the Swedish population norm [25].

To be able to compare to what extent the participants fol-lowed the dietary recommendations we chose three out of seven available questions from the validated Celiac Dietary Adherence Test (CDAT) [26]. One of the questions concerns how hard the patients think it is to eat gluten-free when they are not at home, but we also wanted an estimation of how well they actually managed to stay gluten-free at such times. Therefore, we also asked the patients how often they were able to follow a gluten-free diet when eating outside their home. Answers were graded on a 5-point Likert scale.

Analyses of nominal data, including patient characteristics and information on how the care was organised, were per-formed with Chi-square tests. Normally distributed continu-ous data (based on analysis of skewness) was calculated with the Student’s t-test, and data that was not normally distrib-uted with the Mann-Whitney U-test. Data from the question-naires were ordinal data, hence the Mann–Whitney U-test was used. A p-value <.05 was considered significant in all analyses. Statistical analyses were calculated with IBM SPSS Statistics for Windows, Version 25.0 (Armonk, NY).

The study was approved by the Swedish Ethical Review Authority (protocol number 2019-06068). Written informed consent was obtained from all participating patients.

Results

Of the invited patients 39 out of 73 (53%) agreed toparticipa-tein the GP cohort and 58 out of 121 (48%) in the GE cohort. Those who did not respond were reminded once. All patients had the diagnosis confirmed with biopsy. The mean follow-up time was 7.0 years in the GP grofollow-up and 7.3 years in the GE group (p ¼ .44). The demographic data did not differ between the groups. The most common primary cause of investigation in both groups was abdominal pain. Most patients had some degree of villous atrophy (Table 1).

The participation rate was around 50% despite a reminder. In the group that did not respond 69% were women compared to 64% among the participants and the mean age was 42 years in the drop-out group compared to 43 years. These differences were not significant (p ¼ .45 and .51, respectively). There were no differences within the study sites either.

All patients had a doctor’s appointment at the time of diagnosis. Most patients in the GP group were diagnosed by a primary healthcare physician. In the GE group all patients were diagnosed by a specialist in gastroenterology. While 98% of the patients in GE group received counselling from a registered dietician, only 85% of the patients in the GP

group did (p ¼ .01) (Table 2). In the GE group some labora-tory tests were controlled more often (IgA-tTg, mean corpus-cular volume (MCV), albumin, creatinine and parathyroid hormone (PTH) as well as weight) (Table 3). The patients in the GE group had significantly higher levels of albumin, cre-atinine, PTH and ferritin (Table 4). The follow-up period was also organised differently. Follow-up visits to a physician and dietician were carried out more often in the GE group. Only 31% in the GP group had a follow-up visit to a dietician, to be compared with 93% in the GE group (p .001). In line with this, IgA-tTg serology, laboratory tests and control gas-troscopies were carried out to a larger extent in the GE

Table 1. Baseline characteristics of the study participants.

GP group (n ¼ 39) GE group (n ¼ 58) p-Value Mean (range) Mean (range) Age at diagnosis 42.5 (19–75) 44.1 (19–82) .63 Follow-up time in years 7.0 (5.2–9.9) 7.3 (5.3–10.1) .44

Patients % Patients % Gender Female 27 69 35 60 .37 Cohabitation No, single 9 23 13 22 .94 Yes 30 77 45 78 Smoking Never smoked 24 62 45 78 .05 Quit 12 31 13 22 Current smoker 3 8 0 0 Comorbidities Cardiovascular diseases 9 23 14 24 .90 Pain-related conditionsa 10 26 11 19 .43 Psychiatric diseases 7 18 9 16 .75 IBS and dyspepsia 6 15 5 9 .30 Diabetes mellitus 1 3 3 5 .53 Othersb 11 28 28 48 .05 Primary cause of investigation

Abdominal pain 13 33 16 28 .11 Vitamin deficiencies, anaemia 7 18 14 24

Diarrhoea 7 18 14 24

Weight loss 5 13 0 0

Fatigue 2 5 4 7

Other 5 13 10 17

Histological findings at diagnosis

Intraepithelial lymphocytosis 4 10 6 10 .15 Partial villous atrophy 8 21 24 41 Subtotal villous atrophy 14 36 17 29 Total villous atrophy 13 33 11 19 GP group: Follow-up organised by general practitioners in primary care centres, patients from Malm€o.

GE group: Follow-up organised by gastroenterologists in specialised care, patients from Link€oping.

a_{Chronic pain conditions such as fibromyalgia, arthrosis/arthritis requiring} medication and other conditions requiring analgesics. bThis group primarily includes different allergies, asthma, and skin conditions such as eczema and psoriasis.

The significant results, i.e. those with ap-value .05 or below are highlighted in bold to facilitate the interpretation of the results.

Table 2. Organisation of care at the time of diagnosis. GP (n ¼ 39) GE (n ¼ 58) p-Value Patients % Patients % Doctor_{’s appointment} General practitioner 31 79 0 0 <.001 Gastroenterologist 8 21 58 100 Dietician appointment Yes 33 85 57 98 .01 No 6 15 1 2

(5)

group (Table 5). However, no significant differences were found in Hb, iron status or folate and cobalamin levels 5–10 years after the diagnosis (Table 6).

We asked the patients how they would have wanted the follow-up to be organised if they could choose. Many patients wished to have been followed up by both a special-ist in gastroenterology and a dietician every to every other year (42% in the GP group and 36% in the GE group). In contrast 26% in the GE group and 10% in the GP group thought that they were doing fine by their own, although this was not significant (p ¼ .40) (Figure 1).

The GSRS-20 questionnaire was used to evaluate the pres-ence of gastrointestinal symptoms. We found no evidpres-ence of differences in specific gastrointestinal symptoms when the GSRS dimensions were analysed (Table 7).

The SHS questionnaire was used to evaluate the patients’ subjective health related to their gastrointestinal disease. We found no evidence of any differences between the two cohorts in gastrointestinal symptom burden, social function-ing, disease-related worries and general well-being (Table 8).

To assess if the two groups differed in their experience of fatigue the MFI questionnaire was used. The two cohorts had similar experiences of fatigue (Table 9).

Mental well-being was measured with the PGWB. The median of the total PGWB score was 101 in the GP group and 100 in the GE group (p ¼ .80). Dividing the answers into six sub-scales made no difference and the two cohorts had similar mental well-being (Table 10).

The SF-36 questionnaire gives information both on phys-ical and mental well-being. The patients in the GP group felt more limited in their daily life because of their physical health than the patients in GE group (median 75 and 100, respectively, p ¼ .02). To increase the sensitivity of the analy-ses we chose to use the two sum indexes relating to physical and mental health. The median score on the PCS was 49.7 in the GP group and 52.8 in the GE group (p ¼ .33) and on the MCS 47.7 and 49.6, respectively (p ¼ .92). Hence, the two sum indexes also indicated that the two cohorts did not dif-fer in physical or mental health (Table 11).

Patients self-reported answers to four dietary adherence questions were used to evaluate dietary adherence. In the GP group 28% had eaten gluten-containing food on purpose during the last 4 weeks compared to 9% in the GE group (p ¼ .01) (Figure 2(C)). In addition, the patients in the GP group found it harder to eat gluten-free outside their home (p ¼ .02) and failed more often to eat gluten-free at these

Table 3. Analyses performed at the two sites at the time of diagnosis.

GP (n ¼ 35) GE (n ¼ 58)

p-Value Patients % Patients %

Number of patients where specific analyses were carried out

Weight 14 40 47 81 <.001 IgA-tTg serologya 34 87 58 100 .005 Hb 31 89 55 95 .27 MCV 14 40 54 93 <.001 Albumin 10 29 29 50 .04 Creatinine 19 54 44 76 .03 Iron statusb ₂₁ ₆₀ ₄₁ ₇₁ _.29 Folate 24 69 38 66 .76 Cobalamin 24 69 44 76 .44 PTH 2 6 29 50 <.001 Vit D 2 6 12 21 .05 TSH 16 46 38 66 .06

Abbreviations: IgA-tTg, tissue transglutaminase antibodies; Hb, haemoglobin; MCV, mean corpuscular volume; PTH, parathyroid hormone; Vit D, Vitamin D (25-OH); TSH, thyroid-stimulating hormone.

a_{n ¼ 39 in the GP group.}b_{Assessed with either transferrin saturation or ferritin.}

Table 4. Laboratory values describing coeliac disease-associated morbidity at the time of diagnosis.

GP GE

Mean difference 95% CI p-Value

Mean Mean IgA-tTg (kU/L) 71 68 2.9 17.1 to 22.9 .78 Hb (g/L) 130 133 3.1 10.3 to 4.0 .38 MCV (femtolitre) 84 89 5.2 11.4 to 1.1 .10 Albumin (g/L) 37 42 5.5 10.5 to 0.5 .03 Creatinine (mmol/L) 63 71 8.4 14.4 to 2.3 .008 Iron (mmol/L) 12 14 2.3 6.7 to 2.2 .31 Transferrin saturation 0.18 0.26 0.08 0.2 to 0.001 .05 PTH (pmol/L) 6 22 16.3 26.3 to 6.3 .003 Vit D (nmol/L) 53 53 0.1 32.8 to 32.6 .99 TSH (mIE/L) 2.5 2.1 0.5 0.4 to 1.4 .30

Median (quartiles) Median (quartiles)

Folate (nmol/L) 12 (8–14) 14 (10–20) .08

Cobalamin (pmol/L) 265 (209–375) 248 (216–332) .80

Ferritin (mg/L) 10 (7–72) 46 (14–125) .02

Abbreviations: IgA-tTg, tissue transglutaminase antibodies; Hb, haemoglobin; MCV, mean corpuscular volume; PTH, parathyroid hormone; Vit D, Vitamin D (25-OH); TSH, thyroid-stimulating hormone.

GP group: Follow-up organised by general practitioners in primary care centres, patients from Malm€o. GE group: Follow-up organised by gastroenterologists in specialised care, patients from Link€oping.

(6)

times (p ¼ .04). No difference was found in how important the patients thought it was to avoid unintentional gluten exposure in their diet (Figure 2(A–D)).

Discussion

The main finding of our study was that the structure of fol-low-up did not seem to affect the physical or mental well-being among patients with coeliac disease regardless of whether they attended a GP or a specialist in gastroenter-ology. However, the follow-up in primary care was associated with a somewhat impaired dietary adherence compared to follow-up at a gastroenterology clinic.

While the baseline characteristics of the two cohorts did not differ, there were differences in coeliac disease-related health at the time of diagnosis. The patients in the GP group

had lower levels of albumin and ferritin, which may indicate that they had more severe gastrointestinal inflammation and malabsorption. There was also a tendency that these patients had more severe villous atrophy. Weight loss was a more common cause of investigation in the GP than in the GE group. The patients in the GP group thus seem to have been in a worse condition at the time of diagnosis. This raises the question whether milder cases could be missed in Malm€o. In line with this hypothesis there were 70% more cases of coel-iac disease in Link€oping (GE cohort) compared to Malm€o (GP cohort) during the study (121 cases vs 73) despite the fact that the population is more than twice as big in Malm€o as in Link€oping [27]. At least in childhood coeliac disease, no differ-ence in inciddiffer-ence has previously been reported between the south of Sweden, where Malm€o is situated, and central Sweden, where Link€oping is located [28]. This raises the

Table 5. Organisation of care during follow-up.

GP (n ¼ 39) GE (n ¼ 58)

p-Value Patients % Patients %

Follow-up visit to a doctor

Yes, at least once 23 59 56 97 <.001

No follow-up 16 41 2 3

Provider of care

General practitioner 16 70 0 0 <.001

Gastroenterologist 3 13 51 91

General practitioner and gastroenterologist 4 17 5 9 Number of visits to a doctor

1–2 17 74 27 48 .06

3–4 3 13 22 39

5 or more 3 13 7 13

Follow-up visit to a dietician

Yes, at least once 12 31 54 93 <.001

No follow-up 27 69 4 7

Number of visits to a dietician

1–2 11 92 33 60 .11

3–4 1 8 21 38

5 or more 0 0 1 2

Number of times laboratory tests for deficiencies were carried outa

No tests carried out 8 23 2 3 <.001

Tests carried out 1–2 times 19 54 7 12

Tests carried out regularly 8 23 49 85

Type of laboratory tests

Hb 3 11 11 20 .14

Hb, iron 0 0 2 4

Hb, folate, cobalamin 10 37 9 16

Hb, iron, folate, cobalamin 14 52 34 61

IgA-tTg serology 20 51 45 78 .007

Control gastroscopy 7 18 54 93 <.001

The number of appointments and tests during the entire follow-up period are stated. Abbreviations: IgA-tTg, tissue transglutaminase antibodies; Hb, haemoglobin. a_{n ¼ 35 in the GP group.}

Table 6. Vitamin and mineral deficiency tests 5–10 years after the diagnosis of coeliac disease.

GP GE

Mean difference 95% CI p-Value

n Mean n Mean

Hb (g/L) 24 135 33 138 2.7 8.9 to 3.5 .38

n Median (quartiles) n Median (quartiles)

Folate (nmol/L) 9 15 (12–25) 4 14 (8–22) .44

Cobalamin (pmol/L) 12 321 (253–629) 6 264 (236–357) .19 Ferritin (mg/L) 9 83 (40–273) 8 103 (18–178) .63 Abbreviations: Hb, haemoglobin.

(7)

question if differences in case finding strategies could result in fewer detected cases and also detection at a later stage when the patients are more affected by the disease or if the inci-dence may really differ between different regions in Sweden.

The presence of gastrointestinal symptoms was not affected by the follow-up. It is somewhat surprising that the lack of dietary adherence seen in the GP group did not result in more gastrointestinal symptoms. Though not all of the

items in the GSRS questionnaires relate to typical coeliac dis-ease symptoms, differences in some of the subgroups should have been seen if there indeed were any differences between the two cohorts. The total GSRS-15 sum in both cohorts is in line with previously reported levels of Scandinavian patients on a gluten-free diet for several years [29,30]. To the best of our knowledge, it has not been studied before if the type of healthcare provider affects how much gastrointestinal symptoms patients have. Pekki et al. [31] have at least found that patients who received no fol-low-up at all had more overall symptoms than patients with follow-up for more than 2 years. Consequently, it seems that the follow-up that has been offered in the GP and GE groups 0% 5% 10% 15% 20% 25% 30% 35% 40% 45%

Do not know No follow-up Only physician Only diecian General praconer and diecian Gastrointesnal physician and diecian GP group GE group

Figure 1. The patients’ wishes on how they would have wanted the follow-up care to be organised (p ¼ .40). n ¼ 38 in the GP group, n ¼ 55 in the GE group. GP group: Follow-up organised by general practitioners in primary care centres, patients from Malm€o. . GE group: Follow-up organised by gastroenterologists in speci-alised care, patients from Link€oping.

Table 7. Gastrointestinal symptoms 5–10 years after the diagnosis.

GP GE

p-Value GSRS-15 n Median Quartiles n Median Quartiles

Abdominal pain syndrome 39 2.0 1.0–3.7 58 1.7 1.3–2.7 .53 Reflux syndrome 38 1.0 1.0–2.0 58 1.0 1.0–2.5 .55 Indigestion syndrome 37 2.5 1.6–3.4 58 2.3 1.5–3.0 .55 Diarrhoea syndrome 38 2.0 1.3–3.5 58 1.3 1.3–3.0 .40 Constipation syndrome 39 1.7 1.0–3.0 58 1.7 1.0–3.4 .38 Total score 37 28.0 22.0–48.0 58 30.0 20.8–40.5 .69 GSRS-IBS Pain syndrome 39 2.5 1.0–4.0 58 2.0 1.0–4.0 .71 Bloating syndrome 38 2.7 1.6–4.3 58 2.0 1.6–3.7 .32 Constipation syndrome 39 1.7 1.0–3.0 58 1.7 1.0–3.4 .38 Diarrhoea syndrome 38 2.0 1.3–3.3 58 1.7 1.3–3.0 .70 Satiety syndrome 38 1.0 1.0–3.0 58 1.0 1.0–2.1 .90 Total score 38 27.5 19.0–44.3 58 27.0 18.5–38.3 .59 GSRS, Gastrointestinal Symptom Rating Scale, higher scores indicate more gastrointestinal symptoms. Subdimensions are presented as the sum of the items divided by the number of items in that dimension (range 1–7). The total score is the sum of all items (GSRS-15 range 15–105, GSRS-IBS 13–91).

Table 8.Subjective health related to the patients’ intestinal disease 5–10 years after the diagnosis.

GP GE

p-Value SHS n Median Quartiles n Median Quartiles Symptoms 38 2.0 1.0–4.0 58 2.0 1.0–3.0 .49 Social function 38 2.0 1.0–3.0 57 2.0 1.0–3.0 .58 Disease-related worries 38 2.0 1.0–3.0 58 2.0 1.0–3.0 .93 General well-being 39 2.0 2.0–3.0 58 2.0 2.0–3.0 .97 SHS, Short Health Scale, higher scores indicate more problems because of the gastrointestinal disease (range 1–6).

Table 9. Fatigue 5–10 years after the diagnosis.

GP GE

p-Value MFI n Median Quartiles n Median Quartiles General fatigue 39 12.0 8.0–16.0 58 13.5 7.8–16.3 .62 Physical fatigue 39 10.0 6.0–13.0 58 11.0 6.0–14.3 .70 Mental fatigue 39 8.0 6.0–12.0 58 10.0 6.0–14.0 .16 Reduced motivation 39 8.0 5.0–10.0 58 9.5 6.8–12.0 .05 Reduced activity 39 8.0 5.0–13.0 58 10.0 5.0–14.0 .23 MFI, Multidimensional Fatigue Inventory, higher scores indicate a lot of fatigue (range 4–20).

Table 10. General well-being 5–10 years after the diagnosis.

GP GE

p-Value PGWB n Median Quartiles n Median Quartiles

Anxiety 39 23.0 18.0–27.0 58 24.0 20.0–26.0 .40 Depression 39 16.0 13.0–18.0 58 16.0 13.0–18.0 .39 Well-being 39 17.0 12.0–20.0 58 16.0 11.8–19.0 .75 Self-control 39 16.0 14.0–17.0 57 16.0 13.0–17.0 .77 General health 39 12.0 11.0–17.0 58 13.0 12.0–16.0 .64 Vitality 39 16.0 11.0–20.0 58 15.0 11.0–19.3 .53 Total score 39 101.0 73.0–116.0 57 100.0 81.0–112.5 .80 PGWB, Psychological General Well-being Index, higher scores represent better well-being. Ranges: Anxiety 5–30, depression 3–18, well-being 4–24, self-con-trol 3–18, general health 3–18, vitality 4–24, total score 22–132.

(8)

has been good enough to reduce gastrointestinal symptoms to the same degree.

The general well-being of Swedish coeliac disease patients has previously been studied by Roos et al. [32] in the 1990s. They found a mean PGWB total index of 103 in their cohort, and the same in the general population control group. The median total indexes of our cohorts (101 in the GP group, 100 in the GE group) are close to that. Thus, our study has shown that the general well-being of coeliac disease patients was not influenced by which healthcare provider that was responsible for the follow-up. In line with these findings nei-ther the MFI questionnaire about fatigue nor the SHS (Short

Health Scale) could detect any differences between the two coeliac cohorts.

It is known that initiation of a gluten-free diet generally improves quality-of-life, though not necessarily to the same levels as the healthy general population [7]. The self-reported mental and physical health in the SF-36 questionnaire in both of our cohorts were slightly below the median scores of the general Swedish population (PCS 53.3 and MCS 53.4) [25], though the study was not primarily designed to com-pare the two cohorts with the background population. No differences were found between the two studied groups, except that the patients in the GP group felt more limited

Table 11. Physical and mental health 5_{–10 years after the diagnosis.}

GP GE

p-Value SF-36 n Median Quartiles n Median Quartiles

Physical Functioning 39 100.0 85.0–100.0 58 100.0 85.0–100.0 .90 Role-Physical 38 75.0 0–100.0 58 100.0 50.0–100.0 .02 Bodily Pain 39 72.0 51.0–100.0 58 73.0 41.8–100.0 .59 General Health 39 57.0 47.0–87.0 57 72.0 47.0–87.0 .76 Vitality 39 55.0 35.0–80.0 58 60.0 38.8–76.3 .67 Social Functioning 39 87.5 62.5–100.0 58 87.5 75.0–100.0 .74 Role-Emotional 38 100.0 33.3–100.0 58 100.0 0–100.0 .75 Mental health 39 76.0 56.0–92.0 58 76.0 52.0–89.0 .85 Physical Component Summary 38 49.7 42.2–56.3 57 52.8 46.7–55.9 .33 Mental Component Summary 38 47.7 31.9–53.7 57 49.6 30.0–54.7 .92 SF-36, Short Form 36, higher scores represent better health. Range for the sub-dimensions 0 (worst possible health state) to 100 (best pos-sible health state). Physical Component Summary range 6.8–73.7, Mental Component Summary range 2.7 to 72.1.

0% 10% 20% 30% 40% 50% 60% 70% 80% 90%

Strongly agree Somewhat agree Neither agree nor disagree

Somewhat disagree

Strongly disagree

I am able to follow a gluten-free diet when dining outside of home. GP group GE group 0% 10% 20% 30% 40% 50% 60% 70%

Very important Somewhat important

Neutral/unsure A lile important Not at all important

How important to your health are accidental gluten exposures? GP group GE group (A) (B) (C) (D) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

None of the me A lile of the me

Some of the me

Most of the me All of the me

Over the past 4 weeks, how many mes have you eaten food containing gluten on purpose?

GP group GE group 0% 10% 20% 30% 40% 50% 60% 70% 80% 90%

Always Most of the me Oen Somemes Never

How oen do you usually manage to follow the gluten-free diet when dining outside of home?

GP group GE group

Figure 2. (A) Patients in the GP group found it harder to follow a gluten-free diet outside of home (p ¼ .02). n ¼ 39 in the GP group, n ¼ 56 in the GE group. (B) No difference in how important the participants believed accidental gluten exposure to be (p ¼ .63). n ¼ 39 in the GP group, n ¼ 55 in the GE group. (C) Patients in the GP group ate gluten on purpose more often (p ¼ .01). n ¼ 39 in the GP group, n ¼ 55 in the GE group. (D) Patients in the GP group stayed gluten-free outside of home less often (p ¼ .04). This item is a modified version of the item in (A), in order to get an estimation of how often the patients manage to follow a gluten-free diet outside their homes.n ¼ 39 in the GP group, n ¼ 55 in the GE group. GP group: Follow-up organised by general practitioners in primary care centres, patients from Malm€o. GE group: Follow-up organised by gastroenterologists in specialised care, patients from Link€oping.

(9)

due to their physical health (p ¼ .02). However, there was no difference in the other physical health subdimensions, and we therefore believe that this could be related to mass sig-nificance in the statistical analyses.

A good knowledge of the disease and diet is a key factor to achieve good dietary adherence [33]. The patients in the GP group had fewer visits to physicians and dieticians and did not have their blood samples tested as often as in the GE group. Despite this fact we did not find any differences in Hb, iron status, folate or cobalamin levels at the time of the study, even though it is difficult to draw any conclusions based on that, since it was primarily those already under treatment with vitamin or mineral supplements that were tested. Nevertheless, the dietary adherence was affected by how the patients were followed up. The patients followed up in primary care found it harder to eat gluten-free outside of home, which might reflect an insecurity and lack of know-ledge of the gluten-free diet. In Finland, coeliac disease man-agement has gone through a similar change in organisation as in Malm€o, and most patients are now followed up in pri-mary care. Kurppa et al. [34] have investigated dietary adher-ence in this patient group and concluded that follow-up in primary care could indeed be associated with a good dietary adherence. However, the Finnish study participants were only classified as adherent or non-adherent, and many of them were recruited from coeliac disease advocacy groups. Membership of a coeliac disease advocacy group is known to improve dietary adherence [33]. Kurppa et al. did not find that dietary counselling improved adherence [34]. It has been hypothesised that patients nowadays get enough infor-mation from friends and the Internet [33,34]. Our findings are in contrast to the study by Kurppa et al. [34] and indicate that availability to some kind of counselling could be of value in order to improve dietary adherence in the long run. Maybe it is not the type of healthcare that is offered that makes the difference but instead the quantity and quality of follow-up on an individual basis.

Our study is one of a few that has related health in coel-iac disease patients to how the follow-up has been organ-ised. We have had a broad perspective and evaluated several factors relating to coeliac disease health, including gastro-intestinal symptoms, general well-being, overall mental and physical health as well as dietary adherence. Only patients with a histologically confirmed diagnosis were included. All patient files were scrutinized in order to verify that they met the inclusion criteria, especially concerning age at diag-nosis. It can be argued that patients attending a university hospital (GE group) will suffer from a more advanced disease with complications. However, since no more than one hos-pital covers each study site there are no other ways to follow up the disease than at the hospital or by the general practi-tioners in the region.

However, the study is also limited by a couple of factors. The sample size is unfortunately rather small despite a 5-year inclusion time. As a study like this has not been done before, a relevant power analysis could not be performed before-hand. It was also, as previously discussed, unexpected that the number of cases in Malm€o (GP cohort) would be lower

than in Link€oping (GE cohort), something that could be worthwhile to investigate. Furthermore, the response rate was only around 50% in both of the cohorts. The ongoing COVID-19 pandemic and the rather high number of ques-tionnaires sent out might have affected the response rate.

No major demographic discrepancies could be noted between the participating patients and the drop-out group. However, there were some discrepancies in socioeconomic factors between the two sites’ general populations. At the time of the study, 15% of the population in Link€oping was born abroad vs. 31% in Malm€o. The income levels were e24,000 per year in Link€oping vs. e22,000 in Malm€o. In Link€oping 43% had higher education vs. 41% in Malm€o [27]. Dietary adherence in coeliac disease patients has not been proven to be dependent on most of these factors. Leffler et al. [33] found differences in compliance regarding poverty, knowledge about the gluten-free diet, and membership in a patient association, but no differences in educational level, age, duration on gluten-free diet, autoimmune comorbidities, psychological disorders, symptoms, employment, or marital status. Kurppa et al. [34] did not find the presence of comor-bidities to affect dietary adherence. Consequently, even though the number of articles about these matters are scarce, socioeconomic factors do not seem to influence the dietary adherence, at least not to any major extent. Based on the limited knowledge about the drop-out group the factor that still seems to have the most profound impact on dietary adherence is knowledge about the gluten-free diet.

We chose not to use the CDAT test as a whole but instead selected the items focused on the adherence to a gluten-free diet and excluded those concerning energy level, headaches and anxiety. It can be argued that the whole CDAT should be used but our purpose was to concentrate on the questions regarding the diet. This makes our results less generalizable but nevertheless the outcome gives some information about adherence. The evaluation of dietary adherence could be strengthened by gathering up-to-date IgA-tTg serology tests from all patients or by individual inter-views with dieticians. Further research could focus on gather-ing present-day laboratory values to better be able to evaluate if the organisation of follow-up affects the risk of vitamin or mineral deficiencies.

In conclusion, the adherence to the gluten free diet was better in the group followed at a specialist centre. This could be due to a more structured follow-up with more frequent visits both to physicians and dieticians, which could provide support and knowledge that facilitate the dietary treatment. Despite this, we found that regardless of the design of the follow-up the physical and mental well-being was compar-able 5–10 years after diagnosis. From this respect it could be questioned whether follow-up at a specialist centre is worth-while from a cost-benefit perspective. Regardless of type of follow-up it is crucial that the patients are given adequate and enough dietary counselling to a degree that ensures that their adherence is appropriate. Whether this should be done through dietary counselling by a dietician or in some other way should be studied further. The results of this study could anyway provide an opportunity to improve clinical

(10)

follow-up guidelines of coeliac disease. Follow-up in primary care based on a holistic approach can still be a suitable and equivalent alternative to subspecialised health care if it is well arranged in order to ensure dietary adherence.

Acknowledgements

The authors wish to thank Ingegerd Sundqvist at the Department of Pathology in Malm€o for assistance in retrieving the patient files.

Disclosure statement

No potential conflict of interest was reported by the author(s).

References

[1] Lohi S, Mustalahti K, Kaukinen K, et al. Increasing prevalence of coeliac disease over time. Aliment Pharmacol Ther. 2007;26(9): 1217_–1225.

[2] Rubio-Tapia A, Kyle RA, Kaplan EL, et al. Increased prevalence and mortality in undiagnosed celiac disease. Gastroenterology. 2009;137(1):88–93.

[3] Singh P, Arora A, Strand TA, et al. Global prevalence of celiac dis-ease: systematic review and meta-analysis. Clin Gastroenterol Hepatol. 2018;16(6):823–836.

[4] Ivarsson A, Myleus A, Norstrom F, et al. Prevalence of childhood celiac disease and changes in infant feeding. Pediatrics. 2013; 131(3):e687_–e694.

[5] West J, Fleming KM, Tata LJ, et al. Incidence and prevalence of celiac disease and dermatitis herpetiformis in the UK over two decades: population-based study. Am J Gastroenterol. 2014; 109(5):757–768.

[6] Lebwohl B, Sanders DS, Green PHR. Coeliac disease. Lancet. 2018; 391(10115):70–81.

[7] Cossu G, Carta MG, Contu F, et al. Coeliac disease and psychiatric comorbidity: epidemiology, pathophysiological mechanisms, qual-ity-of-life, and gluten-free diet effects. Int Rev Psychiatry. 2017; 29(5):489–503.

[8] Goddard CJ, Gillett HR. Complications of coeliac disease: are all patients at risk? Postgrad Med J. 2006;82(973):705–712.

[9] Roos S, Wilhelmsson S, Vulcan A, et al. Bowel symptoms, self-image and comorbidity impact on well-being of women with coeliac disease. J Nurs Healthcare Chronic Illness. 2011;3(3): 302_–309.

[10] Bode S, Gudmand-Hoyer E. Symptoms and haematologic features in consecutive adult coeliac patients. Scand J Gastroenterol. 1996; 31(1):54–60.

[11] Kahaly GJ, Frommer L, Schuppan D. Celiac disease and endocrine autoimmunity - the genetic link. Autoimmun Rev. 2018;17(12): 1169–1175.

[12] Kurien M, Trott N, Sanders DS. Long-term care for patients with coeliac disease in the UK: a review of the literature and future directions. J Hum Nutr Diet. 2016;29(5):617_–623.

[13] Silvester JA, Rashid M. Long-term follow-up of individuals with celiac disease: an evaluation of current practice guidelines. Can J Gastroenterol. 2007;21(9):557–564.

[14] Rubio-Tapia A, Hill ID, Kelly CP, et al. ACG clinical guidelines: diagnosis and management of celiac disease. Am J Gastroenterol. 2013;108(5):656–676.

[15] Ludvigsson JF, Bai JC, Biagi F, et al. Diagnosis and management of adult coeliac disease: guidelines from the British Society of Gastroenterology. Gut. 2014;63(8):1210_–1228.

[16] Bai JC, Ciacci C. World Gastroenterology Organisation global guidelines: celiac disease February 2017. J Clin Gastroenterol. 2017;51(9):755_–768.

[17] Sandstr€om O, Agardh D, Ekstav L, et al. Nationellt vårdprogram f€or celiaki. Svensk f€orening f€or Pediatrisk Gastroenterologi, Hepatologi och Nutrition, Svenska barnl€akarf€oreningen, Svensk gastroenterologisk f€orening, Svensk f€orening f€or allm€anmedicin [Internet]. 2020 [cited 2020 Jan 30]. Available from:https://gastro. barnlakarforeningen.se/wp-content/uploads/sites/10/2020/01/ SPGHN_Celiaki_v%C3%A5rdprogram_20200114.pdf

[18] Svedlund J, Sjodin I, Dotevall G. GSRS-a clinical rating scale for gastrointestinal symptoms in patients with irritable bowel syndrome and peptic ulcer disease. Dig Dis Sci. 1988;33(2): 129_–134.

[19] Wiklund IK, Fullerton S, Hawkey CJ, et al. An irritable bowel syn-drome-specific symptom questionnaire: development and valid-ation. Scand J Gastroenterol. 2003;38(9):947_–954.

[20] Ljotsson B, Jones M, Talley NJ, et al. Discriminant and convergent validity of the GSRS-IBS symptom severity measure for irritable bowel syndrome: a population study. United European Gastroenterol J. 2020;8(3):284_–292.

[21] Stjernman H, Granno C, Jarnerot G, et al. Short health scale: a valid, reliable, and responsive instrument for subjective health assessment in Crohn_{’s disease. Inflamm Bowel Dis. 2008;14(1):} 47_–52.

[22] Smets EM, Garssen B, Bonke B, et al. The Multidimensional Fatigue Inventory (MFI) psychometric qualities of an instrument to assess fatigue. J Psychosom Res. 1995;39(3):315_–325.

[23] Wiklund I, Holst J, Karlberg J, et al. A new methodological approach to the evaluation of quality of life in postmenopausal women. Maturitas. 1992;14(3):211_–224.

[24] Sullivan M, Karlsson J, Ware JE. Jr. The Swedish SF-36 Health Survey_{–I. Evaluation of data quality, scaling assumptions,} reliabil-ity and construct validreliabil-ity across general populations in Sweden. Soc Sci Med. 1995;41(10):1349_–1358.

[25] Sullivan M. SF-36 h€alsoenk€at: svensk manual och tolkningsguide [Swedish manual and interpretation guide]. 2nd ed. G€oteborg (Sweden): Sahlgrenska sjukhuset, Sektionen f€or vårdforskning; 2002.

[26] Johansson K, Norstrom F, Nordyke K, et al. Celiac dietary adher-ence test simplifies determining adheradher-ence to a gluten-free diet in swedish adolescents. J Pediatr Gastroenterol Nutr. 2019;69(5): 575_–580.

[27] Statistics Sweden (SCB). Statistical Database [Internet]. Stockholm (Sweden): SCB; 2020 [cited 2020 May 12]. Available from:http:// www.statistikdatabasen.scb.se/pxweb/en/ssd/

[28] Olsson C, Stenlund H, Hornell A, et al. Regional variation in celiac disease risk within Sweden revealed by the nationwide prospect-ive incidence register. Acta Paediatr. 2009;98(2):337_–342. [29] Laurikka P, Salmi T, Collin P, et al. Gastrointestinal symptoms in

celiac disease patients on a long-term gluten-free diet. Nutrients. 2016;8(7):429.

[30] Hallert C, Granno C, Grant C, et al. Quality of life of adult coeliac patients treated for 10 years. Scand J Gastroenterol. 1998;33(9): 933_–938.

[31] Pekki H, Kaukinen K, Ilus T, et al. Long-term follow-up in adults with coeliac disease: predictors and effect on health outcomes. Dig Liver Dis. 2018;50(11):1189_–1194.

[32] Roos S, Karner A, Hallert C. Psychological well-being of adult coeliac patients treated for 10 years. Dig Liver Dis. 2006;38(3): 177_–180.

[33] Leffler DA, Edwards-George J, Dennis M, et al. Factors that influ-ence adherinflu-ence to a gluten-free diet in adults with celiac disease. Dig Dis Sci. 2008;53(6):1573_–1581.

[34] Kurppa K, Lauronen O, Collin P, et al. Factors associated with dietary adherence in celiac disease: a nationwide study. Digestion. 2012;86(4):309_–314.