Anders Beckman

A brief conceptual tutorial of multilevel analysis in social epidemiology: using measures of clustering in multilevel logistic regression to investigate contextual phenomena

Study objective: In social epidemiology, it is easy to compute and interpret measures of variation in multilevel linear regression, but technical difficulties exist in the case of logistic regression. The aim of this study was to present measures of variation appropriate for the logistic case in a didactic rather than a mathematical way. Design and participants: Data were used from the health survey conducted in 2000 in the county of Scania, Sweden, that comprised 10 723 persons aged 18–80 years living in 60 areas. Conducting multilevel logistic regression different techniques were applied to investigate whether the individual propensity to consult private physicians was statistically dependent on the area of residence (that is, intraclass correlation (ICC), median odds ratio (MOR)), the 80% interval odds ratio (IOR-80), and the sorting out index). Results: The MOR provided more interpretable information than the ICC on the relevance of the residential area for understanding the individual propensity of consulting private physicians. The MOR showed that the unexplained heterogeneity between areas was of greater relevance than the individual variables considered in the analysis (age, sex, and education) for understanding the individual propensity of visiting private physicians. Residing in a high education area increased the probability of visiting a private physician. However, the IOR showed that the unexplained variability between areas did not allow to clearly distinguishing low from high propensity areas with the area educational level. The sorting out index was equal to 82%. Conclusion: Measures of variation in logistic regression should be promoted in social epidemiological and public health research as efficient means of quantifying the importance of the context of residence for understanding disparities in health and health related behaviour.

Country of birth, socioeconomic position, and healthcare expenditure: a multilevel analysis of Malmö, Sweden

Study objective: The principle of equity aims to guarantee allocation of healthcare resources on the basis of need. Therefore, people with a low income and persons living alone are expected to have higher healthcare expenditures. Besides these individual characteristics healthcare expenditure may be influenced by country of birth. This study therefore aimed to investigate the role of country of birth in explaining individual healthcare expenditure. Design: Multilevel regression model based on individuals (first level) and their country of birth (second level). Setting: The city of Malmö, Sweden. Participants: All the 52 419 men aged 40–80 years from 130 different countries of birth, who were living in Malmö, Sweden, during 1999. Main results: At the individual level, persons with a low income and persons living alone showed a higher healthcare expenditure, with regression coefficients (and 95% confidence intervals) being 0.358 (0.325 to 0.392) and 0.197 (0.165 to 0.230), respectively. Country of birth explained a considerable part (18% and 13%) of the individual differences in the probability of having a low income and living alone, respectively. However, this figure was only 3% for having some health expenditure, and barely 0.7% with regard to costs in the 74% of the population with some health expenditure. Conclusions: Malmö is a socioeconomically segregated city, in which the country of birth seems to play only a minor part in explaining individual differences in total healthcare expenditure. These differences seem instead to be determined by individual low income and living alone.

Choice of primary care provider: Results from a population survey in three Swedish counties.

Recent reforms in Swedish primary care have involved choice of provider for the population combined with freedom of establishment and privatisation of providers. This study focus to what extent individuals feel they have exercised a choice of provider, why they exercise choice and where they search for information, based on a population survey in three Swedish counties. The design of the study enabled for studying behaviour with respect to differences in time since introduction of the reform and differences in number of alternative providers and establishments of new providers in connection with the reform. About 60% of the population in the three counties felt that they had made a choice of provider in connection with or after the introduction of a reform focusing on choice and privatisation. Establishments of new providers and having enough information increased the likelihood whereas preferences for direct access to a specialist decreased the likelihood of making a choice. The data further suggests that individuals were rather passive in their search for information and tended to choose providers that they previously had been in contact with. This is in line with results from previous studies and poses challenges for county councils governance of reforms.

The role country of birth plays in receiving disability pensions in relation to patterns of health care utilisation and socioeconomic differences: a multilevel analysis of Malmo, Sweden

BackgroundPeople of low socioeconomic status have worse health and a higher probability of being granted a disability pension than people of high socioeconomic status. It is also known that public and private general physicians and public and private specialists have varying practices for issuing sick leave certificates (which, if longstanding, may become the basis of disability pensions). However, few studies have investigated the influence of a patients country of birth in this context.MethodsWe used multilevel logistic regression analysis with individuals (first level) nested within countries of birth (second level). We analysed the entire population between the ages of 40 and 64 years (n = 80 212) in the city of Malmo, Sweden, in 2003, and identified 73% of that population who had visited a physician at least once during that year. We studied the associations between individuals and country of birth socioeconomic characteristics, as well as individual utilisation of different kinds of physicians in relation to having been granted a disability pension.ResultsLiving alone (ORwomen = 1.72, 95% CI: 1.62–1.82; ORmen = 2.64, 95% CI: 2.46–2.83) and having limited educational achievement (ORwomen = 2.14, 95% CI: 2.00–2.29; ORmen = 2.12, 95% CI: 1.98–2.28) were positively associated with having a disability pension. Utilisation of public specialists was associated with a higher probability (ORwomen = 2.11, 95% CI: 1.98–2.25; ORmen = 2.16, 95% CI: 2.01–2.32) and utilisation of private GPs with a lower probability (ORmen = 0.76, 95% CI: 0.69–0.83) of having a disability pension. However, these associations differed by countries of birth. Over and above individual socioeconomic status, men from middle income countries had a higher probability of having a disability pension (ORmen = 1.61, 95% CI: 1.06–2.44).ConclusionThe country of ones birth appears to play a significant role in understanding how individual socioeconomic differences bear on the likelihood of receiving a disability pension and on associated patterns of health care utilisation.

Social inequalities in health- do they diminish with age? Revisiting the question in Sweden 1999

BackgroundIndividuals with low income have poorer health and should, therefore, have higher health expenditure than people with high income particularly in countries with a universal health care system. However, it has been discussed if social inequities in health diminish with age and we, hence, aimed to analyse this question.MethodsWe performed an age stratified cross-sectional analysis using averages, logistic and linear regression modelling of health care contacts, health care expenditures and mortality in relation to individual income in five groups by quintiles. The population consisted of all the 249,855 men aged 40 to 80 years living in the county of Skåne, Sweden during 1999.ResultsFor working-age people (40-59 year old) we find a clear socioeconomic gradient with increasing probability of health care contact, relative expenditure and mortality as income decreased. The point estimations for 1st (highest)-2nd-3rd-4th and 5th (lowest) income groups were: (1.00-1.13-1.21-1.42 and 1.15), (1.00-1.16-1.29-1.69 and 1.89) and (1.00-1.35-1.44-2.82 and 4.12) for health care contact, relative expenditure and mortality respectively. However, in the elderly (75-80 year old) these point estimates were (1.00-0.83-0.59-0.61 and 0.39), (1.00-1.04-1.05-1.02 and 0.96) and (1.00-1.06-1.30-1.33 and 1.49).ConclusionsAs expected among working-age individuals, lower income was associated with higher health care contact, relative expenditure and mortality. However, the existence of opposite socioeconomic gradients in health care utilisation and mortality in the elderly suggests that factors related to a high income might condition allocation of resources, or that current medical care is ineffective to treat determinants of income differences in mortality occurring earlier in the lifecourse.

Many fathers visit the child health care centre, but few take part in parents' groups.

Fathers are expected to play an increasing part in the care of children, but fathers are not automatically included in studies of early childhood. In this paper we present Swedish fathers’ experiences of child health care. In the study, a total of 237 out of 279 (85%) fathers of small children were telephone interviewed using a structured questionnaire. As many as 165 (70%) had visited the child health centre at some time — 91 (38%) regularly or quite often — and we found an association between visits to the child health centre and physician contacts on account of the child’s illness. A common feature throughout was the fathers’ positive view of the child health centre and its activities, with as many as 77 percent grading it as very good or good. Sixty-seven fathers (28%) had taken part in a parents’ group, and here there was a positive association with being a first-time father and knowing about the child’s birth weight and vaccinations. We conclude that professionals must find new ways to get fathers to come to child health care, and especially to parents’ groups.

Changes in health care utilisation following a reform involving choice and privatisation in Swedish primary care: a five-year follow-up of GP-visits

BackgroundThe organisation of Swedish primary health care has changed following introduction of free choice of provider for the population in combination with freedom of establishment for private primary care providers. Our aim was to investigate changes in individual health care utilisation following choice and privatisation in Swedish primary care from an equity perspective, in subgroups defined by age, gender and family income.MethodsThe study is based on register data years 2007 – 2011 from the Skåne Regional Council (population 1.2 million) regarding individual health care utilisation in the form of visits to general practitioner (GP). Health utilisation data was matched with data about individual’s age, gender and family income provided by Statistics Sweden. Multilevel, logistic regression models were constructed to analyse changes in health utilisation in different subgroups and the probability of a GP-visit before and after reform.ResultsHealth care utilisation in terms of both number of individuals that had visited a GP and number of GP-visits per capita increased in all defined subgroups, but to a varying degree. Multilevel logistic regression showed that individuals of both genders aged above 64 and belonging to a family with an income above median had more advantage of the reform, OR 1.25-1.29.ConclusionsReforms involving choice and privatisation in Swedish primary health care improved access to GP-visits generally, but more so for individuals belonging to a family with income above the median.

Effect of vision, proprioception, and the position of the vestibular organ on postural sway.

Abstract Conclusion: When measured together, it seems that vision and proprioception as well as position of the vestibular organ affect postural sway, vision the most. Mediolateral (ML) sway does not seem to be influenced by the position of the vestibular organ. Objective: To investigate how postural sway was affected by provocation of vision, by the position of the vestibular organ, and by provocation of proprioception, when measured together. Methods: Postural sway was measured by using a force plate. Tests were performed with eyes open and eyes closed, with head in neutral position and rotated to the right and to the left and with head maximally extended, both standing on firm surface and on foam. Measures of ML speed (mm/s), anteriorposterior (AP) speed (mm/s), and sway area (SA) (mm2/s) were analyzed using a multilevel approach. Results: The multilevel analysis revealed how postural sway was significantly affected by closed eyes and standing on foam, and by the position of the vestibular organ. Closed eyes and standing on foam both significantly prolonged the dependent measurement, irrespective of whether it was ML, AP or SA. However, only AP and SA were significantly affected by vestibular position, i.e. maximal head movement to the right and extension of the head.

Infection prevention at day-care centres: Feasibility and possible effects of intervention

Objective. To study the effect of an educationally oriented intervention programme, with the recommendations from the National Board of Health and Welfare as a base. Design. A prospective intervention study. Setting. Six day-care centres in Växjö, Sweden. Three centres comprised the intervention group and three constituted the control group. Subjects and main outcome measures. The parents and personnel completed a questionnaire on their views concerning information about infectious diseases. During a nine-month period, parents of all children reported every episode of absence, the number of days absent, the cause of absence, and any contact with doctors or prescription of antibiotics. Results. The guidelines were implementable in routine child day-care. Parents found regular information valuable and felt better informed about infectious diseases. Multilevel analyses showed no statistically significant results of the intervention. “Infection-prone” children had more sickness absence, doctors consultations, and antibiotic prescriptions than those not “infection-prone”. Conclusion. It is possible to implement an educationally oriented intervention programme directed against infectious diseases in child day-care. No significant effect of the intervention was found, which is why a larger intervention study is needed.

Research methods courses as a mean of developing academic general practice

Since 1989, the authors have given courses in research methodology, and these courses are now given at six venues in southern Sweden, as well as in Denmark. The course corresponds to half a years full-time study, with half the time devoted to lectures and studies of literature, while the rest is spent on an individual project under supervision. To enable part-time study, the course extends over 1½ years. In 15 years roughly 1000 people, mainly physicians, have been given training in basic research methods. The course model has been appreciated by clinically active colleagues, who have been able to attend a course and simultaneously work with patients. Among the GPs in the region, one in five has taken this course, and one in five has then gone on to start formal PhD studies. The authors have thus succeeded in their goal of giving basic scientific schooling to many physicians and recruiting some for further research.