Non-participation modestly increased with distance to the examination clinic among adults in Finnish health examination surveys

Aims: Health examination surveys (HES) provide important information about population health and health-related factors, but declining participation rates threaten the representativeness of collected data. It is hard to conduct national HESs at examination clinics near to every sampled individual. Thus, it is interesting to look into the possible association between the distance from home to the examination clinic and non-participation, and whether there is a certain distance after which the participation activity decreases considerably. Methods: Data from two national HESs conducted in Finland in 2011 and 2012 were used and a logistic regression model was fitted to investigate how distance was related to non-participation. Results: We found out that non-participation modestly increased with distance to the examination clinic. An additional analysis indicated that the option of having an examination at home may decrease the effect of distance to participation. Conclusions: Long distances from home to the examination clinic are one reason for low participation activity. Possible bias caused by these differences in participation could be decreased by providing the option of a home examination.


Introduction
Health examinaton surveys (HES) provide informaton about populaton health and health related factors, but declining partcipaton rates threaten the representatveness of collected data.When regional comparisons of health indicators are conducted to investgate possible regional health differences in the study populaton, it is important that the used data are equally representatve in all regions.
Field work in HESs is expensive, so it is hard to conduct natonal HESs so that examinaton clinics would be near to every sampled individual.Especially, this is true in Finland, which is a sparsely populated but geographically relatvely large country.Thus, it is interestng to look into associatons between the distance from home to the examinaton clinic and non-partcipaton, and whether there is a certain distance after which the partcipaton actvity decreases considerably.These findings have relevance when interpretng regional differences in HES results and planning future surveys.
There is no previous literature about the associaton between distance and non-partcipaton in HESs.It has been reported that women with long distances to mammography screening centers have higher nonpartcipaton than those who live close to the centers [1][2][3].

Methods
In these analyses data from two natonal HESs conducted in Finland were used.The Health 2011 study was conducted among adult populaton (29+ years) living in the mainland Finland [4].It was a follow-up study for the Health 2000 study conducted 11 years earlier.The natonal FINRISK 2012 study among adult populaton aged 25-74 years was conducted in five large geographical areas [5].FINRISK 2012 belongs to a series of cross-sectonal FINRISK studies conducted between five years from 1972.Data collecton of the Health 2011 and FINRISK 2012 surveys was based on questonnaires and a health examinaton including physical measures and biological samples.
We restricted the data sets to the common age range 29-74.Individuals who had partcipated in a different examinaton clinic where he/she had been invited were removed from the data (6 individuals).This resulted in the sample sizes of 6816 and 9109 of which 55% and 59% partcipated the health examinaton at examinaton clinics in the Health 2011 study and the FINRISK 2012 study, respectvely.In Health 2011, 4% of invited individuals had the examinaton at home, because it was not possible for them to partcipate at the examinaton clinic.
In both surveys, a sample was drawn from the Natonal Populaton Register.The sample data included informaton about sex, age and geocodes for home.
Logistc regression models were used to model non-partcipaton.Explanatory variables were distance from home to the examinaton clinic, age, sex, informaton whether the individual belonged to the Health 2011 or FINRISK 2012 sample and informaton whether an individual has partcipated in an earlier HES.Distance and age were used as contnuous variables.Restricted cubic splines [6] were used in logistc models to take possible non-linear associatons into account.The selecton of covariates and their possible non-linearites modeled by splines and first-order interactons was based on Bayesian informaton criterion (BIC) [7].The best model included, in additon to the main effects of the above-mentoned variables, interactons of age with sex, study and earlier partcipaton and an interacton of study with earlier partcipaton.Analyses were carried out using the R statstcal software version 3.1.2[8].Distances between homes and examinaton sites were calculated as the shortest routes along the road network using the Digiroad database of the Finnish Transport Agency (see http://www.liikennevirasto.fi/web/en/open-data/digiroad).Dijkstra's algorithm [9] and ArcGIS software [10] were used in the calculaton of the distances.

Results
The sample sizes and partcipaton rates are presented in Table 1.The partcipaton was usually higher among those living close to the examinaton clinics than among those with long distances.Women had higher partcipaton rates than men, older individuals partcipated more actvely than younger and partcipaton was clearly higher among those who had partcipated already in an earlier HESs than among those who had not.The odds rato (OR) of distance per 10 km for non-partcipaton was 1.08 (95% confidence interval (CI): 1.05, 1.11).Thus, non-partcipaton was greater for those living far from the examinaton clinics than for those living close to them.Non-linear associatons between distance and partcipaton were not found, so there does not seem to be a certain distance after which the partcipaton actvity would change notably.Distance was not found to be associated differently with partcipaton among different sexes or in different studies.
Figure 1 illustrates the predicted partcipaton probabilites for distances from 0 to 30km.For a 10km change in distance, absolute changes in probabilites are slightly less than 2 percentage points.We also estmated how partcipaton actvity would have changed if the distances would have been in maximum of 10 km for all individuals invited to the survey.The distances over 10 km were set to 10 km and partcipaton probabilites were predicted using the same model as above.The effect would not have been substantal as the predicted partcipaton rates would have increased from 55% to 56% in Health 2011 and from 59% to 60% in FINRISK 2012.
To analyze how the possibility to have the examinaton at home affected the associaton between distance and partcipaton, we fitted two further models for the Health 2011 data, similarly as for combined data.
When the response was partcipaton at the examinaton clinic, the OR of distance per 10 km for nonpartcipaton was 1.09 (95% CI: 1.05, 1.12) in Health 2011.When partcipaton to home examinaton was also counted as partcipaton, the OR of distance declined to 1.05 (95% CI: 1.02, 1.08).This can be interpreted as an indicaton that providing the possibility for home examinaton could decrease the effect of distance to partcipaton.

Discussion
This study showed that the distance between home and examinaton clinic along the road network modestly predicted non-partcipaton among adults in Finnish HESs.It is not clear if making the network of examinaton clinics denser would be a cost-efficient way to reduce non-partcipaton.This cannot, however, be interpreted so that increasing the distances in these kinds of studies would not cause considerable decrease in partcipaton rates.People who have long distances to examinaton sites are likely to live in areas where they are used to travel long distances from home to work places and services.A limitaton of this study is that we could not take into account the means of transportaton or topographic characteristcs which may substantally affect travel tme.Time spent for travelling might be even more important predictor of partcipaton than the distance.The lack of informaton about the means of transportaton could also be the reason no non-linear associatons were found.Our study also gave an indicaton that a possibility to have an examinaton at home may decrease the effect of distance to partcipaton and thus decrease the possible bias caused by regional differences in partcipaton.

Figure 1
Figure 1 Predicted partcipaton probabilites by sex and study for distances from 0 to 30km.For Health 2011, predictons are for those who have partcipated in an earlier HES and for FINRISK 2012, for those who have not.Age is fixed to 50.

Table 1
Sample sizes (N) and rates for partcipaton at an examinaton clinic (%) in Health 2011 and FINRISK 2012 studies.Ages restricted to 29 -74.