Can non-participants in a follow-up be used to draw conclusions about incidences and prevalences in the full population invited at baseline? An investigation based on the Swedish MDC cohort

Table 1 Background characteristics (%)

	Background population (n = 65,068)	Baseline non-participants (n = 38,594)	Baseline participants (n = 26,474)	Baseline participants, participating in re-examination (n = 21,868)	Baseline participants, not participating in re-examination (n = 4,606)	Substitution approach to reconstruct background population	Extrapolation approach to reconstruct background population
Socio-demographics
Age
50–59	36	38	33	31	39	36	43
60–69	37	36	37	37	36	36	35
70–77	28	26	30	31	26	27	22
Legal sex
Male	40	41	38	38	38	38	38
Country of birth
Sweden	79	73	88	89	83	85	80
Civil status
Married	55	52	61	62	55	57	50
Education
Primary	42	46	36	35	38	37	39
Short secondary	28	26	30	30	30	30	30
Long secondary	11	10	13	13	12	12	12
Tertiary	19	17	22	22	20	21	19
Employment status
Employed	33	31	36	36	37	37	38
Unemployed	8.6	10	5.8	5.6	7.0	6.5	7.9
Sickness absence	3.5	3.9	2.8	2.5	4.0	3.5	4.9
Retired	54	54	55	56	52	53	50
Disposable income
Quintile 1	20	23	16	16	17	16	17
Quintile 2	20	22	17	17	19	18	20
Quintile 3	20	20	20	20	21	20	21
Quintile 4	20	18	23	24	22	23	21
Quintile 5	20	17	24	25	22	23	20
Disease history
Circulatory	16	17	16	16	17	16	18
Diabetes	1.8	2.2	1.2	1.2	1.5	1.4	1.7
Neoplasms	10	9.6	11	11	11	11	11
Respiratory	5.7	6.1	5.1	5.0	5.9	5.6	6.5
Digestive	13	12	13	13	15	14	16
Mental	5.4	7.0	3.1	2.6	5.4	4.5	7.2
MSE			10.7	13.6	4.4	5.4	8.4

Socio-demographics refer to December 31, 2000. Disease history refers to whether the individual was hospitalized for the disease type in question between January 1987 and August 2001. Throughout, individuals who had died or did not live in Sweden by September 1, 2001, were excluded. MSE refers to mean squared error, calculated across all prevalences in the table, in a comparison with the background population

ISSN: 1471-2288