MATERIALS AND METHODS

SAMPLING METHOD

The target population was all non-institutionalised residents ≥ 15 years of age (approximately 45,709,600 people).

A representative sample was obtained with a stratified probabilistic approach using the 1991 census data to determine distribution among the 11 areas of the United Kingdom and the Kish selection method (7) to elect the particular subject to be interviewed within each targeted household. One of eight different selection tables was randomly assigned to a household before the number was called. Based on the number of subjects in the household, their gender and their age, the table indicated which member should be interviewed.

In all, 4972 interviews were completed (79.6% of those approached). The highest rate of completed interviews was in Northern Ireland (86.8%) and the lowest in East Midlands (78.2%; c2:4.019 p<0.05).

Subjects who did not speak English, who suffered from a hearing or speech impairment, or who had an illness precluding the possibility of an interview were excluded.

INSTRUMENT

The interviews were performed by a company specialising in large telephone surveys (BPS Teleperformance, Birmingham) using the Eval Knowledge Based System (8,9), a previously validated, non-monotonic level-2 expert system with a causal reasoning mode (10,11).

The computer program selects the questions and displays them on a monitor. The human interviewer reads each question to the subject and then enters the response.

Depending on the questions, responses are "yes-no", "present-absent-unknown" basis, or a five-point scale and some questions requiring keyboard answers (e.g. name of illness, duration, etc.).

The system is based on a logical reasoning module that allows to administer questions in a manner adapted to the specific individual. Eval pre-emptively eliminates irrelevant questions based upon prior responses. For example, a subject who is thoroughly satisfied with his/her quality of sleep will not be asked about the impact of sleep-related problems.

STATISTICAL ANALYSES

Data from the 1991 UK census of the non-institutionalised population aged 15 years or older were used as the standard population. The weighting procedure used was adjusted for sample design and took into account the geographical distribution of the sample. The unweighted sample included 2894 women and 2078 men, with ages ranging from 15-100 years. After weighting, the sample consisted of 52.2% women and 47.8% men.

These transformation were performed for all variables and the results presented with the weighted percentage. Ninety-five percent confidence intervals (95% C.I.) are also provided for the analysed variables.

Univariate analyses, using Chi square statistics, and multivariate analyses were performed with the SPSS statistical software. Collinearity problems between variables (i.e., information redundancy) were checked. The method of INDICATOR contrasts (12) was used to determine which categories of the independent variables were significantly associated with the presence of snoring and breathing pauses during sleep. Odds Ratios (OR) were calculated according to the different categories with the cut-off point for significance set at 5%.

The International Classification of Sleep Disorders-1990 (ICSD-90) (13) provided the criteria for the identification of sleep disorders.

RESULTS

Forty percent (40.3%, n=2004) of the population reported snoring regularly, men more often than women, with prevalence increasing with age, up to 55 years.

Breathing pauses during sleep, reported by 3.8% (n=190) of the sample, also increased with age (Table 1).

Table 1. Prevalence of Snoring and Breathing Pauses by Sex and Age Group

95% CI = 95% Confidence Interval
a p < 0.05; b p < 0.01; c p < 0.001

	Age groups (years
	15-24 (n=859)	25-34 (n=935)	35-44 (n=855)	45-54 (n=711)	55-64 (n=631)	≥ 65 (n=980)	Total (n=4972)
	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)
Snoring
Total	23.1 (20.3-25.9)	38.1 (35.0-41.2)	45.8 (42.5-49.1)	53.5 (49.8-57.2)	49.3 (45.4-53.2)	37.3 (34.3-40.3)	40.3 (38.1-41.7)
Men	26.1 (22.0-30.2)a	44.9 (40.4-49.4)b	55.0 (50.3-59.7)c	62.0 (57.0-67.0)c	56.5 (51.0-62.0)b	46.8 (41.9-51.7)c	47.7 (45.7-49.7c
Women	20.2 (16.4-24.0)	31.5 (27.3-35.7)	36.8 (32.2-41.4)	45.1 (39.9-50.3)	42.4 (37.0-47.8)	30.9 (27.2-34.6)	33.6 (31.8-35.4)
Breathing pauses
Total	2.5 (1.5-3.5)	2.7 (1.7-3.7))	4.8 (3.4-6.2)	4.6 (3.1-6.1)	5.1 (3.4-6.8)	3.9 (2.7-5.1)	3.8 (3.3-4.3)
Men	1.9 (0.6-3.2)	5.1 (3.1-7.1)c	6.8 (4.4-9.2)b	6.1 (3.6-8.6)a	7.1 (4.2-10.0)a	6.1 (3.7-8.5)b	5.4 (4.5-6.3)c
Women	3.1 (1.5-4.7)	0.4 (0.0-1.0)	2.8 (1.2-4.4)	3 (1.2-4.8)	3.3 (1.4-5.2)	2.3 (1.1-3.5)	2.4 (1.8-3.0)

The association of both snoring and breathing pauses was reported by 2.5% (n=124) of the sample. A total of 7.8% (n=386) of the population did not know whether they snored and/or had breathing pauses during sleep.

There were findings that would be expected from our current knowledge on sleep disordered breathing.

A regression analysis (see table 2) performed on this data obtained an our sample indicate that snoring is significantly associated with being an obese (BMI ≥ 30 mg/m2), married man, aged ≥25 years. It is also significantly associated with daytime sleepiness, napping, nighttime awakenings, high caffeine intake (≥ 6 cups of tea and/or coffee) and smoking.

Table 2: Factors associated to the presence of snoring - Logistic regression models

OR: Odds ratio; 95% CI 95% confidence interval
Categories of reference: Women; 15-24 years old; No medical consultation; BMI ≤ 30; Single; Too long sleep duration; No daytime sleepiness; No napping; Nightmares never; No coffee; no smoking; Never or < 1 time/ week.

Variables	B	Wald	R	OR [95% I.C.]	Sig
Men	0.639	93.615	0.120	1.9 [1.8 - 2.0]	.0000
Age groups
25-34 y.o.	0.360	6.957	0.028	1.4 [1.2 - 1.7]	.0083
35-44 y.o.	0.579	15.900	0.047	1.8 [1.5 - 2.1]	.0001
45-54 y.o.	0.842	31.915	0.069	2.3 [2.0 - 2.6]	.0000
55-64 y.o.	0.764	24.752	0.060	2.1 [1.8 - 2.4]	.0000
≤ 65 y.o.	0.393	6.567	0.027	1.5 [1.2 - 1.8]	.0104
Medical consultations	0.254	11.230	0.038	1.3 [1.1 - 1.4]	.0008
BMI > 30	0.664	34.106	0.071	1.9 [1.7 - 2.2]	.0000
Marital status
Married	0.654	53.636	0.090	1.9 [1.7 - 2.1]	.0000
Sep./ Divorced	0.233	2.870	0.012	1.3 [1.0 - 1.5]	.0902
Widowed	0.140	1.116	0.000	1.1 [0.9 - 1.4]	.2907
Sleep duration
Too short	0.433	4.598	0.020	1.5 [1.1 - 1.9]	.0320
Appropriate	0.451	5.438	0.023	1.6 [1.2 - 1.9]	.0197
Daytime sleepiness	0.171	4.276	0.019	1.2 [1.0 - 1.3]	.0387
Napping
Sometimes	0.226	5.106	0.022	1.3 [1.1 - 1.5]	.0238
At least 2 times/ week	0.160	2.712	0.011	1.2 [1.0 - 1.4]	.0996
Nightmares
< 1 time / month	0.051	0.393	0.000	1.1 [0.9 - 1.2]	.5305
> 1 time / month	0.317	5.854	0.025	1.4 [1.1 - 1.6]	.0155
Coffee
1 or 2 cups / day	0.100	1.444	0.000	1.1 [0.9 - 1.3]	.2295
3 to 5 cups / day	0.144	2.929	0.012	1.2 [1.0 - 1.3]	.0870
≥ 6 cups / day	0.333	9.607	0.035	1.4 [1.2 - 1.6]	.0019
Cigarettes
≤ 20 cigarettes / day	0.289	12.968	0.042	1.3 [1.2 - 1.5]	.0003
21 to 35 cigarettes / day	0.522	5.953	0.025	1.7 [1.3 - 2.1]	.0147
> 35 cigarettes / day	0.316	0.599	0.000	1.4 [0.6 - 2.2]	.4391
Frequency of nightime awakenings
1 time / week	0.110	1.154	0.000	1.1 [0.9 - 1.3]	.2828
> 1 time / week	0.282	12.844	0.041	1.3 [1.2 - 1.5]	.0003

Breathing pauses are similarly significantly associated with being a 35 to 44 years old, man, taking anxiolytics, has being diagnosed with obstructive airway or thyroid disease, and consulted a physician at least once during the last year. Odds ratios are given in table 3.

Table 3: Factors associated to the presence of breathing pauses - Logistic regression models

OR: Odds ratio; 95% CI 95% confidence interval
Categories of reference: Women; 15-24 years old; No obstructive airways diseases; No medical consultation; No anxiety-reducing medication; No thyroid disease; Restorative sleep.

Variables	B	Wald	R	OR [95% I.C.]	Sig
Men	1.062	15.272	0.147	2.9 [2.4 - 3.4]	.0001
Age groups
25-34 y.o.	-0.502	0.656	0.000	0.6 [-0.6- 1.8]	.4180
35-44 y.o.	1.042	4.293	0.061	2.8 [1.8- 3.8]	.0383
45-54 y.o.	0.196	0.110	0.000	1.6 [0.5- 2.7]	.4163
55-64 y.o.	0.461	0.661	0.000	2.1 [1.8 - 2.4]	.0000
≤ 65 y.o.	0.494	0.834	0.000	1.6 [0.6- 2.7]	.3611
Obstructive airways diseases	2.509	40.901	0.252	12.3 [11.5- 13.1]	.0000
Medical consultations	0.774	5.082	0.071	2.2 [1.5- 2.8]	.0242
Anxiety-reducing medication	1.124	3.097	0.042	3.1 [1.8- 4.3]	.0785
Thyroid diseases	2.074	6.583	0.087	8.0 [6.4- 9.5]	.0103
Non restorative sleep	1.033	11.103	0.122	2.8 [2.2- 3.4]	.0009

When snoring and breathing pauses during sleep reported together, there is a significant association with being an obese (BMI ≥ 30 mg/m2; OR= 2.9 [2.3- 3.5]), a man (OR= 4.4 [3.9 - 4.8]), reporting leg pain (OR= 3.1 [2.5- 3.8]), with difficulty to maintain sleep (OR= 2.9 [2.4- 3.4]) and usually does not sleep fully supine (OR= 4.4 [3.1- 5.6).

The model also identifies urinary problems (OR= 3.9 [2.9- 4.8]), high blood pressure (OR= 2.5 [1.8- 3.1]), daytime sleepiness (OR= 2.3 [1.9- 2.7]), daily intake of more than 6 cups of caffeinated beverages (OR= 1.8 [1.2- 2.4]) as significant variables.

DISCUSSION

A nationwide survey as ours, with a large sample is based on interviews responses. To investigate further the responses to the questions, we considered the criteria (A+B+C) outlined in the "International Classification of Sleep Disorders, 1990" that allow diagnosis of obstructive sleep apnea syndrome and investigated the significant independent variables associated with this diagnosis as defined.

The logistic regression model indicated again, on our sample, significant association with:

• being a man (OR= 3.8 [3.3 - 4.2]),
• difficulty maintaining sleep (OR = 4.0 [3.5 - 4.5]),
• daytime sleepiness (3.8 [3.3 - 4.2]),
• high blood pressure (OR = 2.8 [2.2 - 3.5]),
• presence of leg pain (OR = 2.7 [2.1 - 3.3]),
• obesity (BMI ≥ 30 kg/m2 OR =2.0 [1.4 - 2.5]) and
• non restorative sleep (OR = 1.9 [1.4 - 2.3]).

Despite the frequency of sleep related complaints among subjects with snoring and breathing pauses, only 18.2% (n =31) of the subjects with breathing pauses and 9.2% of the snorers (n =185) believed that they had a sleep problem.

Our survey allowed to evaluate the association between the reports of snoring and breathing pauses and three different major health care related problems:

1) DRIVING ACCIDENTS

In our representative sample the UK population 5.3% of the drivers had an accident during the preceding year.

There was, however, no significant difference between snorers (4.6%) subjects with breathing pauses (6.1%) and other subjects (5.9%).

Report of falling asleep at the wheel, however were significantly more frequent in subjects who reported breathing pauses (6.2%) and regular snoring (4.3%) than in other subjects (2.4% c2 8.593; p < 0.05).

2) HEALTH CARE CONSUMPTION

The percentage of regular snorers who had consulted at least once a physician during the past 12 months did not differ significantly from non snorers (62.2% and 60.2%). But there was a significant difference between subjects with breathing pauses and those without (81% vs 60.8% c2 12.385; p < 0.001).

Also health care consumption was significantly higher in subjects reporting breathing pauses during sleep. Thirty one percent of subjects reporting breathing pauses at night sought medical help 6 times or more during the past 12 months compared with only 12% of regular snorers and 11.9% of non snorers (c2 27.013; p < 0.001).

Number of hospitalisation reported by 11% of the sample revealed only a trend due to the too low number of subjects with breathing pauses during sleep. It was reported by 11.4% of snorers and 18.8% of subjects with breathing pauses during sleep.

3) TREATMENT FOR A PHYSICAL ILLNESS NOT LINK BY PHYSICIANS TO A SLEEP RELATED PROBLEM

At time of the interview, 15.5% of the sample was being treated for a physical illness.

This rate was significantly higher in subjects reporting breathing pauses during sleep (39.8%) than in snorers (16.9%) and other subjects (14%; c2 30.384; p < 0.001).

Finally, treated or untreated hypertension was significantly more frequently reported by subjects with breathing pauses during sleep (13.8%) then by snorers (8.6%) and by other subjects (5.6%; c2 21.504; p < 0.005).

4) CALCULATION OF THE PREVALENCE OF OBSTRUCTIVE SLEEP APNEA SYNDROME IN THE UNITED KINGDOM

Based on the criteria available in ICDS - 90 which included presence of daytime sleepiness, the prevalence of OSAS representation sample of the population of the United Kingdom (criteria A+B+C) 1.9%.

The prevalence of OSAS if calculated among subjects between the ages of 35 and 64 years is 1.5% in women (95% C.I. 0.8% to 2.2%) and 3.5% in men (95% C.I. 2.4% to 4.6%).

COMMENTS

Although there have been previous studies of snoring and sleep apnea syndrome in the United Kingdom (14-16) is to our knowledge the first study on a representative sample of the general population.

Some of the findings could be expected from our knowledge of the pathophysiology of obstructive sleep apnea such as the association between presence of obesity or thyroid disease and breathing pauses during sleep. Similarly the survey emphasises the common association between sleep disordered breathing and reports of disrupted nocturnal sleep, non restorative sleep, daytime sleepiness, greater intake of caffeinated beverage and getting drowsy while driving, i.e. sleep disorders.

The survey however indicates that the sleep disorder could be as much labelled "daytime sleepiness" as "insomnia". This latter label may lead to ignore the sleep related breathing problem, and may explain the fact that breathing pauses, during sleep were significantly associated with anxiolytic drug intake.

The fact that obesity is a significant risk factor is not surprising considering the literature and there are now enough valid evidence that have demonstrated that sleep disorders breathing is an independent risk factor for hypertension.

The association between sleep bruxism and sleep disordered breathing included in the logistic regression is a well known clinical observation and seems to be related to the disproportionate maxillo-mandibular anatomy presented by obstructive sleep apnea patients, particularly those with familial aggregates (17).

The prevalence of sleep disordered breathing calculated from the results of the survey is based on the International Classification of Sleep Disorders, a classification system recognised by the World Health Organisation. It provides the diagnostic criteria that were used for the estimates compare to the prevalence of 1.3% (95% = 0.07-0.9%) of OSAS reported in men aged 35 to 65 years based upon a survey performed in 1990 in the town of Wheatly next to Oxford (18), our results are much higher. This can be explained by the criteria selected at that time that probably identify only a severely affected population. Our results are very similar to those reported by Young et al. (19) in the USA (with a prevalence of 2% in women and 4% in men) and by Gislason et al. in Iceland (20).

Our study indicate that OSASD is still widely unrecognised in the British Isles and it also suggest that consumption of health services may be higher in this specific population, raising the question of the cause of this higher health care usage.

REFERENCES

1. Telakivi T, Partinen M, Koskenvuo M, Salmi T, Kaprio J. Periodic breathing and hypoksia in snorers and controls: validation of snoring history and association with blood pressure and obesity. Acta Neural Scand 1987;76:69-75.
2. Lugaresi E, Cirignotta F, Coccagna G, Piana C. Some epidemiological data on snoring and cardiocirculatory disturbances. Sleep 1980;3:221-224.
3. Koskenvuo M, Kaprio J, Telakivi T, Partinen M, Heikkila K, Sarna S. Snoring as a risk factor for ischaemic heart disease and stroke in men. BMJ 1987;294:16-19.
4. Mendelson W. Sleepiness and hypertension in obstructive sleep apnea. Chest 1992;101:903-909.
5. Parish J, Shepard JW. Cardiovascular effects of sleep disorders. Chest 1990;97:1220-1226.
6. Smirne S, Palazzi S, Zucconi M. Chierchia S, Ferini-Strambi L. Habitual snoring as a risk factor for acute vascular disease. Eur Resp J 1993;6:1347-1361.
7. Kish L. Survey Sampling. New York: John Wiley & Sons Inc., 1965.
8. Ohayon M. Sleep Disorders Questionnaire and Decision Trees of the Eval System. Bibliothèque Nationale du Québec- Dépôt légal- ISBN-2-921483-05-X, 1994.
9. Ohayon M. Use of an Expert System (Eval) in Mental Health Epidemiological Surveys. Proceedings of the Twelfth International Congress on Medical Informatics, MIE-94, 1994, 174-179.
10. Ohayon M, Caulet M. Adinfer: experience of an expert system in psychiatry. In Lun KC et al. (Eds), Medical Informatics (MEDINFO 92). Amsterdam: Elsevier Science Publishers, 1992; 615-619.
11. Ohayon M. Validation of a Knowledge Based System (Adinfer) Versus Human Experts. Proceedings of the Twelfth International Congress on Medical Informatics, MIE-94, 1994, 90-95.
12. Hosmer DW. and Lemeshow S., Applied logistic regression. New York: John Wiley & Sons Inc., 1989.
13. Diagnostic Classification Steering Committee, Thorpy MJ, Chairman. International Classification Of Sleep Disorders: Diagnostic And Coding Manual (ICSD). Rochester, Minnesota: American Sleep Disorders Association, 1990.
14. Shapiro CM, Catterall JR, Oswald I, Flenley DC. Where are the British sleep apnoea patients. Lancet 1981;ii:523.
15. Crosby J, Warley A, Stradling JR. Sleep hypoxaemia and its correlates in 480 men aged 35 to 65 years. Thorax 1989;44:353.
16. Stradling JR. Obstructive sleep apnoea and driving. Br Med J 1989;298:904-905.
17. Rojewski TE, Schuller DE, Clark RW, Schmidt HS, Potts RE. Synchronous video recording of the pharyngeal airway and polysomnography in patients with obstructive sleep apnea. Laryngoscope 1981;92:246-250.
18. Stradling JR, Crosby JH. Predictors and prevalence of obstructive sleep apnoea and snoring in 1001 middle-aged men. Thorax 1991;46:85-90.
19. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered breathing among middle-aged adults. N Eng J Med 1993;328:1230-1235.
20. Gislason T, Revnisdottir H, Kristbjarnarson H, Benediktsdottir B. Sleep habits and sleep disturbances amond the elderly - An epidemiological survey. J Intern Med 1993; 234:31-39.