Health and Disease at the Age of Sixty Findings in a Health Survey of 60-year-old Men in Uppsala and a Comparison with Men 10 Years Younger

A health investigation was performed among 331 men aged 60 years in Uppsala. The investigation was performed in the same way as in 50-year-old men (n=2322) in the same community, previously described. Thus it was possible to make certain comparisons between these two populations of middle-aged men. A considerably higher morbidity, glven as point prevalence, was found for diseases of the cardiovascular system and of the endocrine organs, in the older population compared with the younger. Parallel to this a higher consumption of pharmaceuticals was reported by the older men. Only 39% among the older men versus 51 % of the younger men were smokers. Nearly the same number (two-thirds) of men in both age groups were without codable ECG abnormalities in their resting ECG. Among the older men, however, there were more subjects having multiple pathological ECG flndings than in the youngeragegroup. It is concluded that it is possible to reach approximately 80% of the actual population in special health investigations, in middle-aged men. Future studies, preferably in younger subjects, should aim at an early detection and primary prevention of cardiovascular and endocrine diseases.


INTRODUCTION
During recent decades numerous studies have been performed to elucidate factors that increase the risk of cardiovascular disease and death.Cardiovascular disease has been accorded special importance due to its large and increasing contribution to morbidity and mortality in modern society.
In Sweden, studies have been carried out in Gothenburg (34), Stockholm and Uppsala (IS), among other places.This paper deals with general findings in a health survey of 60-year-old men in Uppsala.They were investigated in much the same way as another group of men, at the age of 50 (15) in the same community.These men were born in the period 1920-24.They numbered 2 322 subjects and were investigated in 1970-1973.Comparisons with this younger age group will be made, with respect to blood pressure (37) and glucose tolerance and serum lipid levels.Some additional studies were made on the same population, including the occurrence of cardiac arrhythmias in daily life, alcohol intake and patterns of urinary electrolytes.These will be reported on elsewhere.

MATERIAL
The population consisted of all men born in 1915 living in Uppsala in 1975.They numbered 422.At the time of screening they were living within the same area as were the 50-year-old men of the previous study.
The population of the corresponding area of Uppsala on 1st January 1975 was 108 676.Of the above-mentioned 422 men called for health examination, 331 arrived, which meant a participation rate of 78.4%.
From the registers of the County Council in Uppsala, the names and addresses of all men born in 1915 were collected.In order to get a relevant list, this collection was made 3 months in advance of the forthcoming examination.

METHODS
A letter was sent to each man in the population one week in advance, inviting him to the health survey and explaining its purpose.In the letter the subject was asked to fast and also refrain from smoking from midnight prior to the morning of the investigation.This started at 7.00 am.The examination was performed by the author, with the assistance of the same registered nurse as in the study of 50-year-old men (15).It took place in the outpatient clinic of the Department of Internal Medicine at the University Hospital of Uppsala.
Non-participants were all offered another appointment by direct telephone contact.
The investigation started in the middle of August 1975 and was completed by the end of November the same year.

Personal interview
This procedure included questions about marital and professional status, consumption of medicines, dietary regimens, etc. Contacts with a physician and the reasons therefore were also noted.Each subject was also asked if he took advantage of the general health survey in the county that is offered by the local health authorities every third year.

Blood pressure measurements
Blood pressure (J3P) was measured on the right arm after 10 min in the recumbent position and after another 2 min in the sitting position.The pulse rate was counted immediately prior to the first BP measurement.A mercury wall manometer (Kifa Ercameter) was used.The cuff had a rubber bladder 12 cm wide and 35 cm long.Systolic BP (SBP) and diastolic BP (DBP) were recorded to the nearest 5 mmHg.DBP was recorded at the disappearance of the Korotkoff sounds (phase 5).

Anthropometric measurements
The height without shoes and the weight in undershorts were measured in whole centimetres and kilograms, respectively.In order to define relative body weight, three weight indices were used: 1) an index described by Lind-berg et al. (24), 2) an index based on the findings of insurance holders in the USA (5). 3) Finally the heights and weights of the 50-year-old men in the same community (15) were used as a reference to relative body weight index in the 60-year-old men.Body fat was estimated with a Harpender (9) caliper.Three parts of the body were measured: to the right of the umbilicus, under the angle of the scapula and at the back of the mild-upper arm.All these measurements were made in the sitting position and were recorded in whole millimeters.

Laboratory investigations
All blood samples were taken in the fasting state.The laboratory investigations performed are shown in Table I together with the methods used by our central laboratory.The mean values are also presented in this table.
The morning urine was examined qualitatively for glucose and albumin, using paper sticks.Each participant was then asked to collect all urine for 24 h following screening, in a plastic bottle.
An intravenous glucose tolerance test (IVGTT) was performed in a randomized subgroup of 67 (20.2%) of the subjects who denied diabetes mellitus and all other diseases requiring chronic medication and dietary treatment.These findings will be described separately (38) and will include a comparison towards 50-year-old men concerning insulin secretion at IVGTT.

Electrocardiogram
A resting 12-lead electrocardiogram (ECG) was recorded in all men.This included standard leads I, 11, and 111, unipolar leads aVF, aVL and aVR and finally leads V1-V6, The conventional amplification 1 mV=lO mm was used, with a paper speed of 50 mm/sec.The paper speed and amplification were frequently tested.The ECGs were interpreted by two independent physicians at the Depart-ment of Clinical Physiology according to the Goldmann criteria (13) and to the Minnesota code (31).
A long-term ECG recording covering 6 hours was made (8) but will be discussed elsewhere (23).

Questionnaire and interview
Some results are summarized in Table I1 concerning questionnaire data obtained for the 60-year-old men, as compared with the 50-year-olds.
For the questions concerning angina pectoris (AP) there was a consistently higher percentage of positive replies in the older age group.Only 2.2 % of the younger men reported AP confirmed by a physician, compared with 1 1.5 % in the older group.This five-fold increase was also noted with respect to hospitalization for myocardial infarction.
Only 39.6% of the older men smoked as against 51.0% of the 50-year-olds.However, among the 60-year-old men 13.6% reported that they had stopped smoking after the age of 50.This would mean that 53.2% of those participating in the screening were smokers at the age of 50.Nearly two-thirds (64.9%) of the 60-year-old men who were smokers at the time of screening were not interested in smoking withdrawal trials.
Concerning body weight changes after the age of 30, almost one third (34.2%) of the older men reported an increase in body weight of more than 10 kg after the age of 30.In the younger group 6.6% reported a similar increase in a period of 20 years.
Half (50.8%) of the older men had access to a physician through their employer.Slightly more than three-quarters of the participants (77.8 %) took advantage of the local health examination the last time they were called.In Tables 111 and IV the occurrence of various diagnostic groups and specific diagnoses are shown.
An increase in general morbidity between the ages of 50 and 60 is apparent.Of particular importance, however, is the increase in diseases of the circulatory system and of endocrine and metabolic disorders.
In Table 111 the relative increase in specific diagnosis is seen.A five-fold increase in the prevalence of diabetes mellitus and a four-fold increase in the prevalence of hypertension was found.
This tendency is also reflected by the increase in the use of medical preparations between the ages of 50 and 60.In the younger age group 224 men (9.6%) reported a daily drug intake, compared with 126 (38.1 %) in the older group.The consumption of various classes of drugs in the two age groups is shown in Table V.About 10% of the men born in 1915 were taking diuretics and/or beta blocking agents.No increase was noted in the use of insulin, while the consumption of oral antidiabetic agents was 10 times higher in the older age group.
The higher prevalence of musculoskeletal dis-  orders in the older group is reflected by the greater consumption of analgesics.The more extensive use of hypnotics and sedatives in the older age group should also be noted.

Blood pressure
The distributions of SBP and DBP are shown in Figs. 1 and 2. The mean SBP was 145 mmHg and the mean DBP 87 mmHg in the entire studied population.
In addition to 41 (12.4%) men who were known as hypertensives at the time of screening, 23 (6.9%) were found to have a DBP of 3105 mmHg.Thus the prevalence of hypertension was 19.3 %, comprising the sum of these two population groups.
When the WHO criterion (39) for hypertension was applied, the prevalence increased to 35.0% of the population.had been told previously that their BP was elevated.This should be compared with the other questions in the self-administered questionnaire, where the possible answer "do not know" was given only 12 times altogether concerning previous diseases, stress symptoms etc.A derailed report of the group with hypertension will follow (37).

Anthropometric measurements
The results obtained on application of three weight indices are given in Table VI.
Using the index based on the 50-year-old men in the same community, 59.3 % of the 60-year-old men fell within _+ 10% of the "ideal" weight.When the U.S. insurance index (5) was used, this figure was 59.8%.Using the standards applied by Lindberg et al. (24), however, only 37.8% of the 60-year-old men fell within these limits.

Laboratory investigations
The distributions of the values for erythrocyte sedimentation rate (ESR) and of venous haematocrit are shown in Figs. 3 and 4, respectively.The mean ESR was 8.1 mmlh, and 43 men (13.0%) had a value of I5 mm/h or more.The mean venous haematocrit was 44.9%.Only 15 men (4.5 %) had a haematocrit below 40%.
The values for fasting serum cholesterol are presented in Fig. 5.The mean value was 6.20 mmol/l.Subjects with values of 7.50 mmol/l or higher were re-tested (n=40), and if they exceeded this limit again they were referred to the Department of Geriatrics for further evaluation.
Fig. 6 shows the distribution of serum triglycerides.This shows skewness to the right.However, after logarithmic conversion the distribution was normal (Fig. 7).The mean value was 1.68 mmol/l.Subjects with a value of 2.4 mmol/l or higher ( n =46) were re-tested and followed the procedure mentioned above for serum cholesterol.Finally, it should be mentioned that 36 (1 1 .O %) of the men had detectable amounts of ethyl alcohol in their urine.Urinary electrolytes will be discussed in another article (37).

Electrocardiographic findings
The ECG findings were considered pathological in 52 (15.7%) of the subjects.In a further 19 men (5.7%) the ECG was considered to be possibly pathological.
In Table VII the ECG results, coded according to the Goldmann (13) criteria and to the Minnesota code (31) are shown, together with ECG findings in 50-year-old men in the same community.No codable abnormality was found in 67.1% of the men.The dominating findings were high QRS amplitude, which occurred in 1 1.8 % and QRS axis deviation in 8.5%.

DISCUSSION
The participation rate in this study was somewhat lower (78.4 %) than in the health examination of 50-year-old men in the same community (15), where it reached 81.7%.The somewhat lower figure could be explained by the larger number of older men who were already under the care of another physician.
Another reason for abstaining from the screening might have been access to health controls provided by employers.
In the general health examination provided by the local health authorities, the male participation rate is about 70% for males aged 50 as well as 60 years (19).
In a primary preventive study in Gothenburg in 1970 (38) the participation rate reached 74% for men born in 1915.Almost the same figure was re-ported by Isokoski (20) from a health survey in Finland.In that study the highest participation was found for ages between 35 and 44 years, after which it levelled off.This is in accordance with the findings reported by Napier (10) in their Tecumseh community study.
The questionnaire seemed to be an adequate method of getting information which could in many ways be confirmed by the personal interview.Concerning hereditary questions it has been found (36) that middle-aged men in Uppsala have a good knowledge of their parents' age at and cause of death.As mentioned previously, very few questions were given the answer "do not know" in the self-administered questionnaire.One exception was that 36 (10.9%) of the subjects were uncertain about previous information on a high BP.This is surprising and suggests uncertainty on the part of the physicians handling hypertension rather than anything else.
The comparison between 50-and 60-year-old men shows that the younger age group more often report a high BP in their parents than the older men.This might reflect the fewer opportunities of the older men's parents to be examined by a physician.
The prevalence of chest pain as revealed in the questionnaire seems somewhat higher than the corresponding figures in a study of men aged 55-59 in Finland (30), but is comparable to that reported for 60-to 64-year-old men in Prague (1 1).The validity of the diagnosis AP given in a questionnaire has been discussed by Lundman et al. (25).In this health examination additional information could be obtained at the personal interview in the cases with positive replies to the AP questions.Between 10.9 Studies on smoking habits showed that only 39.6% of the 60-year-old men smoked, as against 51.0% of the 50-year-olds.This diminishing frequency has been discussed by several authors (14, 16,27,35,40).A more negative attitude amongst physicians towards tobacco smoking may have contributed to this age effect.This is evident, in fact, in the group of treated 60-year-old hypertensives, who smoked less than the population in general.
Measurements of haematocrit and ESR have been performed in many screenings, due to the simplicity of these tests and certainly for psychological reasons.However, the diagnostic value of these tests in this screening was not high.
The mean ESR values were 7. 8  (5) and the other on findings in 50-year-old men in Uppsala (15) yielded almost identical results in describing relative weight.When the latter was used there seemed to be more young men with an index of 0.9, and more older men with an index of 1.1.However, these differences were not significant and the proportions of subjects in the two populations with an index between 0.90 and 1.09 were nearly identical.
A comparison of the ECG findings showed that almost the same proportion-approximately twothirds-in the two studied populations had no abnormality according to the Minnesota Code (3 1).
No increase in the occurrence of Q-wave items was found, but proportionately more men in the older group had QRS axis deviation and T-wave items than in the younger men.Atrial fibrillation also showed a higher prevalence in the group of older men.Among the groups with codable abnormalities, there were more men with multiple findings in the older than in the younger age group.
Most reports on ECG findings are based on hospitalized patients, which make epidemiological comparisons difficult.However, in a large material of 122,043 ECGs performed in the U.S. Air Force, Hiss et al. (18) described various findings in different age groups.Proportionately more abnormalities, especially T-wave changes, were registered in higher ages.The same has also been found by Higgins et al. (17) in analysing results from the Framingham study.
The health examination of 60-year-old men gave results that, compared with those for 50-year-old men, may have been influenced by two factors.
Firstly the age factor, which manifested itself as increased morbidity and thus increased consumption of medical preparations.This tendency is of general importance, as the proportion of aged people is going to increase in Sweden.
Two main diagnostic groups-diseases of the circulatory system and diabetes mellitus-merit special interest.Primary preventive activities and early detection of these diseases ought to be given priority in future health screenings of younger groups.
Secondly, factors operating within a changing society may have influenced the results.Thus the decrease in the number of smokers between the ages of 50 and 60, which is a common feature, might be explained in many ways.One contributory factor may be the increased propaganda in the mass media against smoking.In the study of 50-year-old men in the same community (IS), various primary preventive activities were carried out.These included a smoking withdrawal programme, dietary information and treatment of increased BP.Some of these activities may have influenced the attitudes of other groups of middle-aged men.
The relatively larger number of 60-year-old than 50-year-old men who are retired may to a certain extent be ascribed to age, but the more liberal rules as regards retirement, especially on account of medical disablement, probably accounts for some of the increase.
Finally, the expansion of health screening examinations and medical services offered by employers, in particular, must be considered.Thus, it was noted paradoxically that many 60-year-old men who participated in this screening also reported frequent contacts with other medical authorities.Nevertheless, the increased number of 60-year-old men already under the care of other physicians may have been the reason for part of the lack of participation in this health examination.
A self-administered questionnaire modified after Collen (7) was used.It contained 145 questions, including family

Fig. 1 .
Fig. 1.Distribution of supine systolic blood pressure in the total population.

Fig. 3 .
Fig. 3. Distribution of erythrocyte sedimentation rates in the total population.

Fig. 6 .
Fig. 6.Distribution of serum triglyceride values in the total population.Arithmetically plotted.

Fig. 7 .
Fig. 7. Distribution of serum triglyceride values in the total population.Geometrically plotted.

Table I . Laboratory investigations performed in the health investigation of 60-year-old m e n
Mean values, standard deviations and laboratory methods used by the laboratory

Table 11 .
Percentage of positive replies to some questions in the sev-administered questionnaire, in health surveys of 50-and 60-year-old men in Uppsala

Table 111 .
Prevalence of diagnoses, reported as spec8c diagnosis, in health surveys of 50-and 60-year-old men in Uppsala

Table VII .
Frequency of some codable ECG abnormalities according to the Minnesota Code in health surveys of 50-and 60-year-old men in Uppsala 5% in the older age groups had a haematocrit below 40 %.In the older group none of the men had a haematocrit below 40% without a known cause.The mean haematocrit was slightly higher in the older age group.The indices used in describing relative weight showed that the Norwegian based index rqcommended by Lindberg et al. (24) classified relatively more men as overweight in both the 50and 60-year age groups than the other two indices used.This is probably explained by the relatively lean Norwegian population chosen as a reference for the former weight index.The two other indices used, one based on findings in the Build and Blood Pressure study in USA