Comparative study of characteristics and disease management between subjects with frequent and occasional gastro-oesophageal reflux symptoms

Summary

Background  Little is known about the distinctive characteristics of subjects with frequent (at least weekly) and occasional gastro-oesophageal reflux symptoms.

Aim  To compare the characteristics and disease management of subjects complaining of at least weekly and less frequent gastro-oesophageal reflux symptoms.

Methods  Population-based postal survey carried out in France in 2003 among a representative sample of 8000 subjects.

Results  The prevalence of frequent and occasional gastro-oesophageal reflux symptoms was 7.8% and 23.4%, respectively. Compared to subjects with occasional gastro-oesophageal reflux symptoms, those with frequent symptoms were older, suffered from more severe symptoms and felt greater impact on daily living, despite a slightly shorter duration of symptoms. These subjects more often sought medical advice. Most of them had treated the last episode of symptoms primarily with a proton-pump inhibitor and less often with antacids/alginates. The degree of treatment satisfaction was lower in subjects with frequent gastro-oesophageal reflux symptoms, in relation to a more frequently observed persistence of symptoms irrespective of the medication used except for proton-pump inhibitors.

Conclusions  This survey suggests that subjects complaining of frequent or occasional gastro-oesophageal reflux symptoms constitute two distinctive groups. Despite greater healthcare use, the former group shows a lower level of satisfaction with disease management. Nevertheless, a substantial subset of subjects with occasional symptoms also complained of impaired health-related quality of life and sought health care.

Introduction

Although there is no ‘gold standard’ for the diagnosis of gastro-oesophageal reflux disease (GERD),¹ it has been suggested that the frequency of gastro-oesophageal reflux symptoms (GORS) needs to be considered when deciding whether GORS constitute a disease.² To establish the prevalence of GERD, most epidemiological studies refer to at least weekly symptoms. However, less frequent symptoms could be sufficient to impair the individual's health-related quality of life, so that occasional symptoms might also result in disease for a subset of these subjects. Unfortunately, few data are available in population-based studies about these subjects as well as about differences between those with frequent and occasional symptoms.^{3, 4} Locke et al.³ assessed risk factors associated with frequent and occasional GORS, but they did not compare both groups of patients. More recently, Diaz-Rubio et al.⁴ reported in a Spanish population aged 40–79 years that frequent GORS were directly linked to a psychosomatic symptom score of >8 and inversely associated with educational level and coffee consumption. However, these studies gave no information about the use of medical care in both groups of subjects. In 2003, we conducted a large population-based survey to provide the first estimate of GORS prevalence in France.⁵ Data from this survey afforded for the first time the opportunity to compare the characteristics and disease management of subjects complaining of at least weekly and less frequent GORS.

Methods

This was a postal survey carried out by the public opinion poll institute TNS Sofres from 30 June 2003 to 8 August 2003, from the Access Santé polling base, representative of ordinary households living in the French metropolitan area. The survey questionnaire was mailed to a sample of 8000 subjects aged 15 or over, selected at random. It consisted of 46 questions relating to subject attributes, lifestyle factors, medical history, reflux-related symptom characteristics (frequency, severity, impact on daily living), consultation behaviour (motivation, frequency, information provided, satisfaction), previous treatments for GORS and description of the last episode (medical consultation, medications used, persistence of symptoms, satisfaction with the treatment).

Subjects were asked if they suffered from heartburn and acid regurgitation. Heartburn was described as a burning feeling rising from the stomach towards the neck. Acid regurgitation was described as a sour or bitter liquid coming into the mouth. GERD was classified as frequent if symptoms occurred at least weekly. Symptoms were defined as ‘frequent’ if they occurred at least weekly and ‘occasional’ if they occurred less frequently during the last year. Symptom severity was assessed on a 4-point scale (from not severe at all to very severe), impact on daily living on a scale of 0–10 (from no discomfort at all to very troublesome) and quality of information provided by the physician consulted on a scale of 0–10 (from no information received to feeling very well informed). Body mass index (BMI) was calculated according to Quetelet's formula;⁶ overweight was considered as BMI ≥25 and obesity as BMI ≥30. Alcohol consumption was considered moderate if up to three units per day (approximately 30 g of alcohol).

Numerical data were described by the usual statistics (mean, confidence interval at 95% [95% CI]). Values were adjusted for sex and age. Categorical data were described by the numbers and relative proportions. Between-group statistical comparisons were carried out using the Z-test for categorical data (if the numbers were greater than 30) or a Student's t-test for numerical data. Statistical significance was considered at P < 0.05.

Results

Prevalence of GORS (total, frequent, occasional)

Of the 8000 subjects who were contacted, 5597 (70.0%) responded. As 202 questionnaires had to be discarded (blank, incomplete or incoherent), 5395 subjects (67.4%) effectively participated in the survey. Of these subjects, 1681 had experienced heartburn and/or acid regurgitation, giving a prevalence of GORS of 31.3% (95% CI 30.0–32.4%): 10% of the subjects had experienced only heartburn, 14% only acid regurgitation and 7% both typical GORS. The prevalence of frequent GORS was found in 7.8% (95% CI 7.1–8.5%) (419 subjects/5395) and that of occasional GORS in 23.4% (95% CI 22.3–24.5%) (1262 subjects/5395). Ten questionnaires from subjects with occasional GORS could not be assessed further and were excluded from the analysis.

Of the 419 subjects with frequent symptoms, 23% experienced symptoms once a week, 45% two or three times per week and 32% every day. Among the 1252 subjects with occasional symptoms, 22% experienced symptoms at least monthly (one to three times per month) and 78% less frequently. Most subjects with frequent GORS than occasional GORS suffered from heartburn (77% vs. 49%, P < 0.01) and fewer from acid regurgitation (63% vs. 70%, P < 0.05). Most subjects with frequent GORS experienced both typical symptoms (41% vs. 19%, P < 0.01).

Comparison of characteristics of subjects with frequent or occasional GORS

The main sociodemographic attributes of subjects with frequent or occasional GORS are presented in Table 1. The sex ratio did not differ significantly between the two groups of subjects. Compared to subjects with occasional GORS, there were more subjects with frequent GORS aged 50 years or older, and fewer of those who were professionally active were middle executives, managers or members of the liberal professions. As regards risk factors, elevated BMI, alcohol consumption and current smoking were not more frequent among subjects with frequent GORS.

Table 1.  Sociodemographic, physical, lifestyle and medical attributes of subjects with frequent gastro-oesophageal symptoms (GORS) and occasional GORS

	Subjects with frequent GORS (n = 419)	Subjects with occasional GORS (n = 1252)
Values were adjusted for sex and age. The values in brackets are 95% confidence intervals. *P < 0.01, †P < 0.05, vs. subjects with less frequent GORS. ‡ Overweight, body mass index ≥25 and <30; obesity, body mass index ≥30. § Defined as three or more alcohol units (≥30 g alcohol) per day. ¶ As graded on a 4-point verbal scale (not severe at all, mild, severe, very severe). ** As assessed on a scale of 0–10. †† Asthma, bronchitis/chronic obstructive pulmonary disease, cardiac disorder, and/or otorhinolaryngological disorder. ‡‡ In subjects who had consulted (360 subjects with frequent GORS; 541 subjects with occasional GORS).
Female (%)	54 (49–59)	56 (54–59)
Age 50 years or older (%)	55* (51–60)	41 (38–44)
Profession (%)
Farmers	1 (0–2)	1 (1–2)
Craftsmen, shopkeepers	4 (2–6)	3 (2–4)
Managers, liberal professions	6 (3–8)†	9 (7–10)
Middle executives	10 (7–13)	12 (11–14)
Employees	16 (13–20)	17 (15–19)
Workers	18 (15–21)	16 (14–18)
Retired	28 (25–32)	27 (25–29)
Body mass index mean	25.9 (25.5–26.4)	25.7 (25.4–25.9)
Overweight/obesity ‡ (%)	35 (30–39)/16 (12–19)	34 (31–37)/15 (13–17)
Alcohol consumption§ (%)	7 (5–9)	6 (4–7)
Current smoking (%)	18 (14–21)	21 (19–23)
GORS history duration years	8.1 (7.3–8.9)*	9.8 (9.2–10.3)
Severe or very severe symptoms ¶ (%)	37 (34–40)*	5 (3–7)
Impact on daily living** mean	5.7 (5.4–5.9)*	2.6 (2.5–2.8)
Associated disorders††(%)	29 (25–33)*	23 (21–25)
Medical consultation
At any point in life (%)	85 (81–89)*	44 (42–47)
During the last 12 months (%)	58 (54–62)*	21 (18–23)
Number of consultations during
The last 12 months‡‡ mean	3.3 (3.0–3.6)*	1.9 (1.7–2.2)

Gastro-oesophageal reflux symptom duration was slightly shorter in subjects with frequent GORS (Table 1). However, a greater proportion of these subjects considered their symptoms as severe or very severe than subjects with occasional GORS. Similarly, the impact of GORS on daily living was scored significantly higher by subjects with frequent GORS: 10% of subjects with frequent GORS gave a score of 9 or 10 when compared with only 1% of those with occasional GORS. More subjects with frequent GORS complained of associated disorders (asthma, bronchitis/chronic obstructive pulmonary disease, cardiac and/or otorhinolaryngological disorders).

Medical consultation

Compared with subjects with occasional GORS, most subjects with frequent GORS had consulted a physician for GORS at any point in life (85% vs. 44%, P < 0.01) or during the last 12 months (58% vs. 21%, P < 0.01). The mean number of consultations during the last year was higher in those with frequent GORS (3.3 vs. 1.9) (Table 1). Subjects with frequent GORS were more likely to consult a general practitioner alone (35% vs. 28%, P < 0.01), a gastroenterologist alone (11% vs. 5%) or both of them (36% vs. 9%, P < 0.01).

Subject profiles in terms of medical consultation were compared in both populations (Table 2). In both groups, subjects who had consulted were more frequently 50 years old or older and professionally inactive, had more severe symptoms and felt greater impact of GORS on daily living. Compared with the corresponding subjects with occasional GORS, those with frequent GORS who had consulted were less often female, more frequently over 50 years old and had severe symptoms markedly more frequently.

Table 2.  Sociodemographic, medical attributes and consultation behaviour attributes of subjects with frequent gastro-oesophageal symptoms (GORS) and occasional GORS

	Subjects with frequent GORS (n = 419)		Subjects with occasional GORS (n = 1252)
	Had consulted (n = 360)	Had not consulted (n = 59)	Had consulted (n = 541)	Had not consulted (n = 697)
Values were adjusted for sex and age. The values in brackets are 95% confidence intervals. * P < 0.05, † P < 0.01 vs. subjects who had not consulted. ‡ P < 0.05, § P < 0.01 vs. corresponding subjects with occasional GORS. ¶ As assessed on a scale of 0–10. ** In subjects who delayed consulting for GORS (251 subjects with frequent GORS, 354 subjects with occasional GORS).
Female (%)	53 (48–58)	59 (47–72)	61 (57–65)*	52 (49–56)
Mean age	52.6 (50.9–54.2)*	46.3 (42.1–50.4)	50.9 (49.5–52.2)†	43.2 (42.1–44.4)
Age 50 years or older (%)	58 (53–63)	42 (30–55)	51 (47–55)†	33 (30–37)
Professionally inactive (%)	46 (42–51)	37 (26–47)	43 (39–46)	40 (37–43)
GORS history duration years	8.3 (7.4–9.2)‡	6.9 (4.7–9.1)	9.9 (9.1–10.6)	9.6 (8.9–10.4)
Severe symptoms (%)	41 (38–45)†§	12 (5–20)‡	10 (7–12)†	1 (0–4)
Impact on daily living¶ mean	5.9 (5.7–6.2)†§	4.1 (3.5–4.7)§	3.5 (3.3–3.7)†	1.9 (1.8–2.1)
Previous endoscopy (%)	57 (54–61)§	–	31 (28–34)	–
Reasons for consulting for GORS (%)
Reasons relating to symptoms:	82 (78–87)§	–	56 (52–60)	–
Symptoms too frequent	49 (45–54)§		22 (18–26)
Symptoms hardly bearable	44 (40–49)§		20 (16–24)
Impact on daily life	25 (20–29)‡		19 (15–22)
Symptoms too painful	29 (24–33)§		17 (14–20)
Desire to know the cause	<1 (0–1)‡		2 (1–3)
Appearance of serious symptoms	1 (0–2)		1 (0–2)
Counselling from:
relations	18 (14–22)		19 (16–22)
The pharmacist	4 (2–6)		6 (4–8)
Fear of a serious disease	16 (12–19)		16 (13–19)
GORS mentioned during a consultation for another medical condition	2 (0–4)§		7 (6–9)
Delay to first consultation for GORS (%):
Less than 1 month	27 (22–31)	–	28 (25–32)	–
More than 1 month:	69 (65–74)		66 (62–70)
1 to 3 months	21 (17–25)		20 (17–24)
4 to 6 months	16 (13–20)‡		12 (9–14)
7 to 12 months	7 (4–10)		8 (5–10)
More than 12 months	25 (21–30)		26 (23–30)
Reasons for delaying consultation for GORS (%)**:
Opinion that the disorder was not serious	50 (44–56)	–	51 (46–56)	–
Opinion that the disorder would be transient	42 (36–47)		37 (32–42)
Self-medication	22 (16–27)		26 (22–30)
Counselling by the pharmacist	7 (4–10)		6 (3–8)
Symptoms were not disturbing	9 (4–13)§		17 (13–20)
Reasons for not consulting for GORS (%):
Opinion that symptoms were not serious	–	54 (42–66)‡	–	67 (64–71)
Symptoms not troublesome		16 (3–28)§		33 (30–37)
Self-medication		58 (47–69)§		23 (20–26)
Counselling by the pharmacist		14 (7–20)‡		6 (4–7)
Symptoms infrequent		0 (0–5)		4 (3–5)
Pregnancy		0 (0–4)		2 (1–4)
Knowledge of the cause of the symptoms		5 (1–9)		2 (1–3)

To clarify the relationship between consultation behaviour, frequency of GORS and age, we compared the consultation behaviour among subjects older or younger than 50 (Table 3). In both age groups medical consultation was more frequent in subjects with frequent than with occasional GORS. Age had no effect on the consultation behaviour in subjects with frequent GORS. Patients with occasional GORS had consulted more frequently when older than 50 years (54% vs. 36% for consultation at any point in life, 27% vs. 15% for consultation during last year, P < 0.01).

Table 3.  Consultation behaviour according to age and frequency of gastro-oesophageal reflux symptoms (GORS)

Consultation behaviour	Subjects <50 years		Subjects >50 years
	Frequent GORS (n = 187)	Occasional GORS (n = 737)	Frequent GORS (n = 232)	Occasional GORS (n = 515)
* P < 0.01 vs. occasional GORS. † P < 0.01 vs. subjects >50 years.
Had consulted (%)	153 (82)*	266 (36)†	207 (89)*	276 (54)
Had not consulted (%)	34 (18)*	471 (64)†	25 (11)*	239 (46)
Had consulted during the last year (%)	100 (53)*	111 (15)†	144 (62)*	139 (27)

Among subjects who had consulted, those with frequent GORS had more often been referred for further investigation (60% vs. 35%), especially for endoscopy (58% vs. 31%), when compared with subjects with occasional GORS (P < 0.01 for all differences).

In both groups of subjects who had consulted, reasons for consultation were mainly related to symptoms (frequency, pain, impact on daily living), and the delay to first consultation was very similar (Table 2). In both groups, the reasons for delaying consultation were chiefly the opinion that the condition was not serious and that the symptoms would be transient, and self-medication (Table 2). Among subjects who had not consulted, the two main reasons for not consulting were, for subjects with frequent GORS, self-medication and the opinion that the condition was not serious and, for subjects with occasional GORS, the opinion that the condition was not serious and that symptoms were not troublesome (Table 2).

Most subjects deemed ‘very satisfying’ the information provided by physicians on symptoms (mean score of 7.3–7.4 on the scale of 0–10 according to subject group) and on therapy (mean score of 7.2–7.3). However, most subjects with frequent GORS preferred additional information (36% vs. 24% of those with occasional GORS, P < 0.01).

Therapeutic management of GORS

Most subjects with frequent GORS declared having previously used a medication for the treatment of their symptoms (93% vs. 57% of those with occasional GORS, P < 0.01). Compared to subjects with occasional GORS, most subjects with frequent GORS had used a drug therapy to treat their symptoms during the last 12 months (89% vs. 43%, P < 0.01). The therapy that was used had more often been administered continuously (38% of subjects vs. 4%, P < 0.01) and had more often included a prescription treatment (76% vs. 28%, P < 0.01). The drugs that were used had been mainly proton-pump inhibitors (PPIs) (54% of subjects vs. 11%, P < 0.01) and/or antacids/alginates (47% vs. 20%, P < 0.01).

Therapeutic management of the last episode of GORS Of the 1102 subjects who had previously used a drug therapy for GORS (389 subjects with frequent GORS and 713 with occasional GORS), 838 (76.0%) had undergone treatment for the last episode: 331 subjects with frequent GORS and 507 subjects with occasional GORS (85% vs. 71%, P < 0.01).

Subjects with frequent GORS had more often used a prescribed treatment (80% vs. 47% of subjects with occasional GORS, P < 0.01), less often a self-medication (23% vs. 48%, P < 0.01) and less often a medication advised by the pharmacist (7% vs. 12%, P < 0.01). The profiles of subjects with frequent or occasional GORS who had used a prescription treatment (alone or combined) were compared to those of subjects who had only used a non-prescribed therapy (Table 4). In both populations the severity of symptoms and the impact on daily living were significantly higher in subjects who had received a prescription treatment than in those using a non-prescribed treatment (Table 4).

Table 4.  Sociodemographic and medical attributes of subjects who had treated the last episode of symptoms with either a prescription treatment (alone or combined) or a not prescribed therapy alone

	Subjects with frequent GORS (n = 331)		Subjects with occasional GORS (n = 499)
	Prescription treatment (n = 264)	Not prescribed therapy alone (n = 67)	Prescription treatment (n = 240)	Not prescribed therapy alone (n = 259)
Comparison between subjects with frequent gastro-oesophageal symptoms (GORS) and those with occasional GORS. Values were adjusted for sex and age. The values in brackets are 95% confidence intervals. * P < 0.01, † P < 0.05 vs. subjects who used a not prescribed therapy alone. ‡ As assessed on a scale of 0–10. § P < 0.01 vs. corresponding subjects with less frequent GORS.
Female (%)	54 (48–60)	57 (45–69)	64 (57–70)	54 (48–60)
Mean age	54.4 (52.5–56.5)	48.8 (45.0–52.7)	53.2 (51.1–55.2)*	46.7 (44.8–48.7)
Age 50 years or more	63 (57–69)	49 (37–60)	57 (51–64)*	42 (36–48)
Professionally inactive (%)	46 (41–51)	42 (33–52)	46 (41–52)†	37 (32–42)
GORS history duration years	8.3 (7.2–9.3)	8.2 (6.2–10.3)‡	9.2 (8.1–10.4)†	11.6 (10.5–12.8)
Severe symptoms (%)	46 (42–51)*§	20 (11–28)§	13 (8–17)†	4 (0–8)
Impact on daily living‡ mean	6.1 (5.8–6.4)*§	4.7 (4.1–5.2)§	3.9 (3.6–4.2)*	3.0 (2.7–3.3)

Overall, most subjects with frequent GORS had used PPIs (57% vs. 21% of subjects with occasional GORS, P < 0.01) and fewer had used antacids/alginates (60% vs. 71%, P < 0.01), whereas the proportion of subjects who had used prokinetics (<15%) or histamine H₂-receptor antagonists (<10%) was similar in both groups. Fewer subjects with frequent GORS had used treatments in monotherapy (64% vs. 71% for subjects with occasional GORS, P < 0.01) and more of them had used combinations (34% vs. 20%, P < 0.01).

The prescription treatments used by subjects with frequent GORS had more frequently been prescribed by gastroenterologists (21% of subjects vs. 9% of those with occasional GORS, P < 0.01). Treatments had been mainly prescribed continuously (44% of subjects with frequent GORS vs. 12% of those with occasional GORS, P < 0.01), less often intermittently (23% vs. 37%, P < 0.01) or on demand (22% vs. 40%, P < 0.01). Prescription treatments used by subjects with frequent GORS consisted mainly of PPIs (69% of subjects vs. 37% of those with occasional GORS, P < 0.01) or antacids/alginates (46% vs. 55%, N.S.), less often of histamine H₂-receptor antagonists (6% vs. 11%, P < 0.05) or prokinetics (16% vs. 15%, N.S.).

Persistence of symptoms after treatment of the last episode Symptoms had persisted in 24% of subjects with frequent GORS vs. 12% of those with occasional GORS (P < 0.01). Among subjects with frequent GORS the proportion of those suffering from severe symptoms was higher in the group with persistent symptoms than in the group with no persistence (55% vs. 37%, P < 0.01), but the impact on daily living was not significantly different between both groups (6.1 vs. 5.7, N.S.). Twenty to 30% of subjects with frequent GORS complained of persistent symptoms irrespective of the treatment used while antacids/alginates yielded the best results for those with occasional GORS. In both groups the persistence of symptoms on PPI treatment was more frequently reported when PPIs were used in combination rather than in monotherapy (Table 5).

Table 5.  Persistence of symptoms after treatment of the last episode in subjects with frequent or occasional gastro-oesophageal reflux symptoms (GORS) in relation to the treatment used

	Subjects with frequent GORS		Subjects with occasional GORS
	Number % of subjects	% with persistence	Number of subjects	% with persistence
* P < 0.01, ** P < 0.05 vs. subjects with less frequent GORS. † P < 0.01 vs. the relevant counterparts who had used a treatment in monotherapy. ‡ P < 0.05 vs. treatment on demand. § Low number of subjects.
Overall symptom persistence	331	24 (19–29)*	507	12 (9–15)
Treatments used
Proton-pump inhibitors (PPIs)	183	28 (22–35)	106	22 (14–31)
Antacids/alginates	201	28 (23–33)*	359	9 (5–13)
Prokinetics	51	30 (19–42)	60	18 (7–30)
Histamine H2-receptor antagonists	18§	20 (3–36)	42	11 (0–21)
In monotherapy	214	19 (15–24)*	360	10 (7–14)
PPIs	93	17 (10–25)	56	16 (6–27)
Antacids/alginates	101	22 (15–28)*	268	9 (5–13)
In combination	111	35 (27–43)*†	101	16 (8–24)
On prescription
PPIs in monotherapy	102	21 (13–29)	57	21 (10–32)
PPIs in combination	74	41 (30–52)†	34	30 (14–47)
Mode of administration prescribed
Continuous	113	26 (18–34)	31	26 (10–41)‡§
Intermittent	61	36 (25–46)**	87	21 (12–31)
On demand	57	25 (16–35)*	96	10 (3–17)

Satisfaction with the treatment used for the last episode of GORS The proportions of subjects expressing complete or moderate satisfaction were 67% and 30%, respectively, for subjects with frequent GORS vs. 80% and 17% for those with occasional GORS (P < 0.01 for both between-group differences). Fewer subjects with frequent GORS than occasional GORS expressed complete satisfaction when the treatment had been a prescription treatment (64% vs. 75% respectively, P < 0.01).

The level of complete satisfaction on treatment with PPIs was unaffected by symptom frequency (68% vs. 77% in subjects with frequent or occasional GORS respectively). In contrast, the level of complete satisfaction on treatment with antacids/alginates was significantly lower in subjects with frequent GORS (63% vs. 83%, P < 0.01).

Discussion

This survey is the first to provide a thorough comparison of the characteristics of subjects with frequent and occasional GORS. Data from this survey indicate that these two groups of subjects differ significantly from each other on most analysed items. Indeed, compared to subjects with occasional GORS, those with frequent GORS show a different distribution of typical symptoms (more frequent heartburn, less frequent acid regurgitation, heartburn and acid regurgitation more frequently occurring together), an elevation of perceived levels of symptom severity and impact on daily living, an increased healthcare use, an altered pattern of therapeutic management and lower efficacy of some treatments as assessed by symptom persistence and satisfaction with GORS treatment.

The overall prevalence of GORS in France is 31.3%, i.e. within the range of 26–60% observed for other Western countries. In their literature review, Moayyeddi and Axon⁷ reported that the prevalence of frequent GORS defined as occurring at least once per week varied in Western populations between 8% and 18% with a pooled rate of 12%. The value of 7.8% that we found for the prevalence of frequent GORS in the present survey is slightly lower. As expected however, the prevalence of occasional GORS in the present survey was found to be three times higher than that of frequent GORS (23.4% vs. 7.8%).

Subjects with frequent GORS in the present survey were more often aged 50 or older than those with occasional GORS. The relationship between GORS and age is controversial. Results from a US survey³ suggested that age was inversely associated with GORS regardless of frequency, although the association with frequent symptoms was of borderline significance while no association was found in a Spanish study.⁴ The increase in associated disorders noted in the present survey in subjects with frequent GORS may be linked to the more advanced age in this group of subjects.

In the present survey, as in the Spanish study,⁴ gender does not seem to exert any influence on GORS frequency. Education level and an elevated psychosomatic symptom score have also been found to be inversely and directly related, respectively, to symptom frequency.⁴ A lower proportion of upper-level education subjects among those with frequent GORS was also observed in the present survey. GORS history duration appears to be slightly shorter in subjects with frequent GORS than in those with occasional GORS. This observation contradicts the hypothesis that there might be a transition of GERD subjects from occasional to frequent GORS with time. The present cross-sectional survey could not, however, address this issue which can only be investigated in a long-term longitudinal study.

In accordance with other studies,^{4, 8, 9} perceived GORS severity increased markedly with symptom frequency. Impact on daily living was found to increase in parallel with symptom severity in all survey groups (by symptom frequency) and subgroups (by consultation behaviour or type of treatment used) of patients. Such a close relationship between GORS severity and impact on activities of daily living explains why the impact on daily activities or productivity at work¹⁰ could serve to define GORS severity. With regard to risk factors, 47–51% of subjects with GORS were found to be overweight/obese in the present survey, a frequency slightly greater than that observed in the French general population (42%),¹¹ but there was no significant difference between subjects with frequent or occasional GORS. A BMI of 25 or over is considered to be a strong independent risk factor of GORS.¹² US and Spanish surveys^{3, 4} demonstrated a relationship between being overweight or obese and increased risk of GORS, but with no clear influence on symptom frequency. As in the US survey,³ excess alcohol intake seems slightly more common among subjects with frequent GORS, but the difference between groups was not statistically significant. Smoking unexpectedly seems as common among subjects with occasional as in those with frequent GORS. It may be speculated that GORS may have caused changes in smokers’ lifestyle, inducing them to abstain from tobacco.

The easy accessibility of the public healthcare system in France could explain the high percentage of subjects who have consulted for GORS. Subjects with frequent GORS consulted a physician more frequently during the last year than those with occasional GORS (58% and 20% respectively): the proportion consulting both a general practitioner and a gastroenterologist was four-fold greater (36% vs. 9%). These figures are greater than those reported in other countries.^{3, 4, 13, 14} The reasons for seeking medical advice are known to be related to GORS characteristics (frequency, severity and nocturnal symptoms) or concern about the clinical significance of the symptoms.^{7, 13, 15} This study clearly establishes the relationship between consultation behaviour, symptom frequency and age. While age had no effect on the consultation behaviour in subjects with frequent GORS, more patients especially those over 50 with occasional GORS had consulted. Irrespective of the frequency of GORS, we observed that the severity of symptoms and the impact on daily activities were quite different between consulters and non-consulters. Nevertheless, it is noteworthy that 12% of subjects with frequent GORS who did not consult had severe symptoms. Louis et al.⁹ noted that one-fifth of patients complaining of heartburn with a substantial negative effect on daily activities had not sought medical advice. The higher severity of symptoms does not appear to shorten the delay to first consultation for GORS, which remained greater than 1 month for two-thirds of subjects and greater than 1 year in a quarter regardless of symptom frequency. It is also noteworthy that self-medication remains an important factor of non-consultation for subjects with frequent GORS (57%).

Among subjects who had consulted, most subjects with frequent than occasional GORS had undergone endoscopy (58% vs. 31%). When considering the whole population, 49.8% and 13.4% of subjects with frequent and occasional GORS, respectively, had had an endoscopy. This difference may be partly related to the greater proportion of subjects aged 50 years and older (58% vs. 51%), but more probably to the greater proportion of subjects with severe symptoms (41% vs. 10%), especially those with frequent GORS. Regardless of GORS frequency, these figures are markedly higher than those recently reported from Olmsted County, MN, USA.¹⁶ In the US survey, the authors observed that only 19% of those with frequent GORS in the community had had an upper endoscopy over a 10-year period.¹⁶ Differences between both studies could be related to the fact that in France subjects have open access to gastroenterologists and endoscopy. In a previous observational study we noted that 74% of patients who consulted a gastroenterologist in France underwent an upper endoscopy.¹⁷

As expected, most subjects with frequent than occasional GORS usually treated the symptoms (93% vs. 57%). For the treatment of GORS, subjects with frequent GORS took more prescription treatments (more frequently used continuously), fewer treatments in monotherapy and more combinations. When subjects used prescription treatments, the proportion of those with frequent GORS who used PPIs was nearly two-fold of that noted among subjects with occasional GORS (69% vs. 37%). On the other hand, among subjects who used self-medication or drugs advised by the pharmacist, the proportion of those who took antacids/alginates was similar (>80%) in the two groups of subjects regardless of GORS frequency.

After treatment, symptoms tended to persist more often in subjects with frequent GORS than in those with occasional GORS. Age and GORS duration did not appear to significantly influence symptom persistence, as opposed to symptom severity and type of treatment taken. It is noteworthy that the rate of persistent symptoms was quite similar irrespective of the treatment used in those with frequent GORS. This result contrasts with the established superiority of PPIs for the control of GORS compared with other drugs. These figures may suggest that treatments were prescribed in accordance with heterogeneous clinical presentations. Similarly, the persistence of symptoms – more frequently reported by patients taking PPIs in combination than those using monotherapy – could suggest that a subset of subjects is resistant to any therapy regardless of the frequency of GORS.

The fact that, in this study, satisfaction with treatment was assessed by the subjects themselves reinforces the value of the results as recent studies have indicated that physicians generally tended to underestimate GORS as well as patients’ expectations and need for improvement of symptom control.^17–19

In conclusion, this is the first survey to describe the differences existing between subjects with frequent and occasional GORS. The population with frequent GORS exhibits more severe symptoms and more pronounced impact on daily living. These subjects show an increased healthcare use. However, a subset of subjects with occasional GORS also experiences a substantial impact on daily activities and seeks medical advice.

Conflict of interest

A. Caekaert and P. Barthélemy are employees of AstraZeneca, France.

Acknowledgements

This study was sponsored by AstraZeneca, France.