What is the effect of independent medical evaluation on days on sickness benefits for long-term sick listed employees in Norway? A pragmatic randomised controlled trial, the NIME-trial.

BACKGROUND: Independent medical evaluations are used to evaluate degree and reason for work disability, uncertainty around the functional status, and/or the employee's rehabilitation potential in several jurisdictions, but not in Norway. The main aim of this trial was to test the return to work effect of independent medical evaluation (IME) (summoning and consultation) compared to treatment as usual (TAU) in Norway, for workers who have been on continuous sick leave for 6 months. METHODS: This was a pragmatic randomised controlled trial including all employees aged 18-65 years, sick-listed by their general practitioner and on full or partial sick leave for the past 26 weeks in Hordaland County, Norway in 2015/16. Trial candidates were drawn from a central register at the Norwegian Labour and Welfare Administration at 22 weeks of sick leave. Pregnant women, individuals with cancer or dementia diagnoses, those with secret address, employed by NAV or sick listed by the specialist health services were excluded. Separate regression analyses were conducted to investigate the "intention-to-treat" and "treatment on the treated" effects, using the ordinary least squares and instrumental variable methods, respectively. RESULTS: After exemption based on predefined exclusion criteria, 5888 individuals were randomised to either IME (n = 2616) or TAU (n = 2599). The final intervention group constitutes 1698 individuals, of which 937 attended the IME consultation. No baseline differences were found between the IME and TAU group regarding gender, age, and previous sick leave. Individuals attending the IME were older than those who cancelled the appointment ((47/45), p = 0.006) and those who did not show up without cancelling ((47/42), p < 0.001). Mainly the IME physician agreed with the regular GP upon level of sick leave. In cases with different assessments, the difference tended to be towards a lower sick leave level. There were no intention to treat or treatment on the treated effect on days of sick leave after randomisation during follow up. CONCLUSIONS: Overall, the analyses showed no effect of IME on changes in sick leave for sick listed employees. This result was consistent for those who were offered an IME consultation (intention to treat) and those who undertook an IME consultation (treatment on the treated). TRIAL REGISTRATION: ClinicalTirals.gov trial number NCT02524392 first registration 14.08.2015.

Continuity of care, measurement and association with hospital admission and mortality: a registry-based longitudinal cohort study.

OBJECTIVE: To assess whether continuity of care (COC) with a general practitioner (GP) is associated with mortality and hospital admissions for older patients We argue that the conventional continuity measure may overestimate these associations. To better reflect COC as a GP quality indicator, we present an alternative, service-based measure. DESIGN: Registry-based, population-level longitudinal cohort study. SETTING: Linked data from Norwegian administrative healthcare registries, including 3989 GPs. PARTICIPANTS: 757 873 patients aged 60-90 years with ≥2 contacts with a GP during 2016 and 2017. MAIN OUTCOME MEASURE: All-cause emergency hospital admissions, emergency admissions for ambulatory care sensitive conditions, and mortality, in 2018. RESULTS: We assessed COC using the conventional usual provider of care index (UPCpatient) and an alternative/supplementary index (UPCGP list) based on the COC for all other patients enlisted with the same preferred GP.For both indices, the mean index score was 0.78. Our model controls for demographic and socioeconomic characteristics, prior healthcare use and municipality-fixed effects. Overall, UPCGP list shows a much weaker association between COC and the outcomes. For both indices, there is a negative relationship between COC and hospital admissions. A 0.2-point increase in the index score would reduce admissions for ambulatory care sensitive conditions by 8.1% (CI 7.1% to 9.1%) versus merely 1.9% (0.2% to 3.5%) according to UPCpatient and UPCGP list, respectively. Using UPCGP list, we find that mortality is no longer associated with COC. There was greater evidence for an association between COC and all-cause admissions among patients with low education. CONCLUSIONS: A continuity measure based on each patient's contacts with own preferred GP may overestimate the importance of COC as a feature of the GP practice. An alternative, service-based measure of continuity could be suitable as a quality measure in primary healthcare. Facilitating continuity should be considered a health policy measure to reduce inequalities in health.

Health effects of reduced workload for older employees.

To keep elder employees in the labour force, introducing age-dependent job conditions can be a policy measure. However, we know little about the effect of such initiatives. We investigate the effects of a particular programme in Norway that reduces the workload of teachers at age 55 but maintains the same wage. Evaluation of this programme is well suited to a difference-in-difference analysis, where the control group is teachers slightly too young to be eligible for the workload reduction. Using full population register data for the period 2006-2013, we analyse the effects of the programme on health as indicated by sickness absence and health care utilization. We find that whereas there is no effect among women, the workload reduction causes a decrease in sickness absence and an improvement in mental health among males. These results, which are robust to a placebo test, to extending the pretreatment period, and to dropping single birth cohorts, are driven by a subgroup of men whose prior health status is poor.

Competition and physician behaviour: Does the competitive environment affect the propensity to issue sickness certificates?

Competition among physicians is widespread, but compelling empirical evidence on its impact on service provision is limited, mainly due to endogeneity issues. In this paper we exploit that many GPs, in addition to own practice, work in local emergency centres, where the matching of patients to GPs is random. The same GP is observed both with competition (own practice) and without (emergency centre). Using high-dimensional fixed-effect models, we find that GPs with a fee-for-service (fixed-salary) contract are 12 (8) percentage points more likely to certify sick leave at own practice than at the emergency centre. Thus, competition has a positive impact on GPs' sicklisting that is strongly reinforced by financial incentives.

Socio-economic status and physicians' treatment decisions.

This paper studies the relationship between patients' socio-economic status and general practitioners' (GPs') service provision by exploiting administrative patient-level data with information on consultation length, medical tests, and fee payments for each visit in Norway over a 5-year period (2008-2012). To reduce patient heterogeneity, we limit the sample to a given condition, diabetes type II, that is treated almost exclusively in primary care. We estimate GP fixed-effect models and control for a wide set of patient characteristics. Our results show that, for each visit, patients with low education get shorter consultations but more medical tests, patients with low income get less of both, and patients with low education/income get less services in monetary terms. We also find that, during a year, patients with low education/income visit the GP more often and receive more services in monetary terms. Thus, GPs treat patients differently according to their socio-economic status, but we find no support for a social gradient.

Protocol for the effect evaluation of independent medical evaluation after six months sick leave: a randomized controlled trial of independent medical evaluation versus treatment as usual in Norway.

BACKGROUND: It has been discussed whether the relationship between a patient on sick leave and his/her general practitioner (GP) is too close, as this may hinder the GP's objective evaluation of need for sick leave. Independent medical evaluation involves an independent physician consulting the patient. This could lead to new perspectives on sick leave and how to follow-up the patient. METHODS/DESIGN: The current study is a randomized controlled trial in a Norwegian primary care context, involving an effect evaluation, a cost/benefit analysis, and a qualitative evaluation. Independent medical evaluation will be compared to treatment as usual, i.e., the physicians' and social insurance agencies' current management of long-term sick-listed patients. Individuals aged 18-65 years, sick listed by their GP and on full or partial sick leave for the past 6 months in Hordaland county will be included. Exclusion criteria are pregnancy, cancer, dementia or an ICD-10 diagnosis. A total sample of 3800 will be randomly assigned to either independent medical evaluation or treatment as usual. Official register data will be used to measure the primary outcome; change in sickness benefits at 7, 9 and 12 months. Sick listed in other counties will serve as a second control group, if appropriate under the "common trend" assumption. DISCUSSION: The Norwegian effect evaluation of independent medical evaluation after 6 months sick leave is a large randomized controlled trial, and the first of its kind, to evaluate this type of intervention as a means of getting people back to work after long-term sickness absence. TRIAL REGISTRATION: ClinicalTrials.gov NCT02524392 . Registered June 23, 2015.

Prioritization and the elusive effect on welfare - a Norwegian health care reform revisited.

The Faster Return to Work (FRW) scheme that Norwegian authorities implemented in 2007 is an example of a policy that builds on the human capital approach. The main idea behind the scheme is that long waiting times for hospital treatment lead to unnecessarily long periods of absence from work. To achieve a reduction in average sickness absence duration, the allocation of FRW funds and new treatment capacity is exclusively aimed at people on sick leave. Many countries have allocated funds to reduce waiting times for hospital treatment and research shows that more resources allocated to the hospital sector can reduce waiting times. Our results support this as the FRW scheme significantly reduces waiting times. However, on average the reduction in waiting times is not transformed into an equally large reduction in the sickness absence period. We find significant difference in the effects of FRW on length of sick leave between surgical and non-surgical patients though. The duration of sick leave for FRW patients undergoing surgical treatment is approximately 14 days shorter than for surgical patients on the regular waiting list. We find no significant effect of the scheme on length of sick leave for non-surgical patients. In sum, our welfare analysis indicates that prioritization of the kind that the FRW scheme represents is not as straightforward as one would expect. The FRW scheme costs more than it contributes in reduced productivity loss. We base our analyses on several different econometric methods using register data on approximately 13,500 individuals over the period 2007-2008.

Diffusion of pharmaceuticals: cross-country evidence of anti-TNF drugs.

This article studies the diffusion of biopharmaceuticals across European countries, focusing on anti-TNF drugs, which are used to treat autoimmune diseases (e.g., rheumatism, psoriasis). We use detailed sales information on the three brands Remicade, Enbrel and Humira for nine European countries covering the period from the first launch in 2000 until becoming blockbusters in 2009. Descriptive statistics reveal large variations across countries in per-capita consumption and price levels both overall and at the brand level. We explore potential sources for the cross-country consumption differences by estimating several multivariate regression models. Our results show that large parts of the cross-country variation are explained by time-invariant country-specific factors (e.g., disease prevalence, demographics, health care system). We also find that differences in income [gross domestic product (GDP) per capita] and health spending (share of GDP) explain the cross-country variation in consumption, while relative price differences seem to have limited impact.

The impact of different prioritisation policies on waiting times: case studies of Norway and Scotland.

We investigate the distributional consequences of two different waiting times initiatives, one in Norway, and one in Scotland. The primary focus of Scotland's recent waiting time reforms, introduced in 2003, and modified in 2005 and 2007, has been on reducing maximum waiting times through the imposition of high profile national targets accompanied by increases in resources. In Norway, the focus of the reform introduced in September 2004, has been on assigning patients referred to hospital a maximum waiting time based on disease severity, the expected benefit and the cost-effectiveness of the treatment. We use large, national administrative datasets from before and after each of these reforms and assign priority groups based on the maximum waiting times stipulated in medical guidelines. The analysis shows that the lowest priority patients benefited most from both reforms. This was at the cost of longer waiting times for patients that should have been given higher priority in Norway, while Scotland's high priority patients remained unaffected.

Between two beds: inappropriately delayed discharges from hospitals.

Acknowledging the necessity of a division of labour between hospitals and social care services regarding treatment and care of patients with chronic and complex conditions, is to acknowledge the potential conflict of interests between health care providers. A potentially important conflict is that hospitals prefer comparatively short length of stay (LOS) at hospital, while social care services prefer longer LOS all else equal. Furthermore, inappropriately delayed discharges from hospital, i.e. bed blocking, is costly for society. Our aim is to discuss which factors that may influence bed blocking and to quantify bed blocking costs using individual Norwegian patient data, merged with social care and hospital data. The data allow us to divide hospital LOS into length of appropriate stay (LAS) and length of delay (LOD), the bed blocking period. We find that additional resources allocated to social care services contribute to shorten LOD indicating that social care services may exploit hospital resources as a buffer for insufficient capacity. LAS increases as medical complexity increases indicating hospitals incentives to reduce LOS are softened by considerations related to patients' medical needs. Bed blocking costs constitute a relatively large share of the total costs of inpatient care.

Interdependency between social care and hospital care: the case of hospital length of stay.

BACKGROUND: Poor co-ordination and collaboration have been identified by many governments as a major and growing weakness of their health care systems. Better integrated care for the elderly individuals is one field of particular importance. In this study, we ask to what extent local authorities' social care services create cost externalities by prolonging hospital length of stay (LOS) because of inadequate service capacity and/or service quality. METHODS: The data set is constructed by merging in-patient data from the Norwegian Patient Register with Statistics Norway's local authority variables for the period from 2007 to 2009. The sample includes ∼386 000 observations of in-patients aged ≥ 67 years. Using the quantile regression (QR) technique, we analyse the impact of social care services along the entire distribution of LOS. The QR estimates are compared with ordinary least square estimates (OLS). Patient variables in the analyses include age, gender and case-mix variables. Hospital and time-fixed effects are also controlled for variables. RESULTS: More resources to the social care services give shorter LOS, and the QR analysis shows that resources matter more for patients in the long tail of the distribution compared with those in the lower quantiles. LOS is longer for patients with change of residence after discharge from hospital compared with those patients that do not change residence. CONCLUSIONS: Increased supply of social care services contributes to a reduction in aggregate societal costs of treatment and nursing of elderly patients by shortening comparatively costly hospital LOS.

Monitoring prioritisation in the public health-care sector by use of medical guidelines. The case of Norway.

This paper presents a new way to monitor priority settings in public health-care systems. We take departure in medical guidelines prescribing acceptable waiting times for different medical descriptions. Allocating ICD10 codes to the medical descriptions, we are able to compare actual waiting times to the recommended maximum waiting times. This way we use the medical guidelines as a tool for monitoring prioritisation in the health sector. In an application, using data from the Norwegian Patient Register, we test statistically for compliance with the guidelines. The results indicate that patients suffering from the most severe conditions are receiving too low priority in the Norwegian health-care sector relative to patients of lower priority.

Does variation in general practitioner (GP) practice matter for the length of sick leave? A multilevel analysis based on Norwegian GP-patient data.

In Norway, as in many countries, the national insurance system is under economic stress from demographic change impacting on the pensions versus contributions balance, and an increasing number of disability and sickness benefit claimants. The general practitioner (GP) is responsible for assessing work capacity and issuing certificates for sick leave based on an evaluation of the patient. Although many studies have analyzed certified sickness absence and predictive factors, no studies assess its variation between patients, GPs or geographical areas within a multilevel framework. Using a rich Norwegian matched patient-GP data set and employing a multilevel random intercept model, the study attempts to disentangle patient, GP and municipality-level variation in the certified sickness absence length for Norwegian workers in 2003. We find that most observed patient and GP characteristics are significantly associated with the length of sick leave (LSL) and medical diagnosis is an important observed factor explaining certified sickness durations. However, 98% of the unexplained variation in the LSL is attributed to patient factors rather than influenced by variation in GP practice or differences in municipality-level characteristics. Our findings indicate that GPs practice variation does not matter much for the patients' LSL. Our results are compatible with a high degree of patient involvement in current general practice. Based on this understanding one may infer that GPs play an advocate role for their patients in Norway, where the patients' own wishes are important when decisions are made.

Prioritization and patients' rights: analysing the effect of a reform in the Norwegian hospital sector.

The right to equal treatment, irrespective of age, gender, ethnicity, socio-economic status and place of residence, is an important principle for several health care systems. A reform of the Norwegian hospital sector of 2002 may be used as a relevant experiment for investigating whether centralization of ownership and management structures will lead to more equal prioritization practices over geographical regions. One concern was variation in waiting times across the country. The reform was followed up in subsequent years by some other policy initiatives that also aimed at reducing waiting lists. We measure prioritization practice by a method that takes departure in recommended maximum waiting times from medical guidelines. We merge the information from the guidelines with individual patient data on actual waiting times for the period 1999-2005. This way we can monitor whether each patient in the available register of actual hospital visits has waited shorter or longer than what is considered medically acceptable by the guideline. The results indicate no equalization between the five new health regions, but we find evidence of more equal prioritization within four of the health regions. Our method of measuring prioritizations allows us to analyse how prioritization practice evolved over time after the reform, thus covering some further initiatives with the same objective. The results indicate that an observed reduction in waiting times after the reform have favoured patients of lower prioritization status, something we interpret as a general worsening of prioritization practices over time.

Access to primary health care and health outcomes: the relationships between GP characteristics and mortality rates.

This paper analyses the impact of economic conditions and access to primary health care on health outcomes in Norway. Total mortality rates, grouped into four causes of death, were used as proxies for health, and the number of general practitioners (GPs) at the municipality level was used as the proxy for access to primary health care. Dynamic panel data models that allow for time persistence in mortality rates, incorporate municipal fixed effects, and treat both the number and types of GPs in a district as endogenous were estimated using municipality data from 1986 to 2001. We reject the significant relationship between mortality and the number of GPs per capita found in most previous studies. However, there is a significant effect of the composition of GPs, where an increase in the number of contracted GPs reduces mortality rates when compared with GPs employed directly by the municipality.

A panel data study of physicians' labor supply: the case of Norway.

Physicians are key personnel in a sector which is important due to its size as well as the quality of service it provides. We estimate the labor supply of physicians employed at hospitals in Norway, using personnel register data merged with other public records. A dynamic labor supply equation is estimated using a sample of 1303 male physicians observed over the period 1993-1997. The methods of estimation are GMM and system GMM. We reject the static model in favor of a dynamic model and obtain short run wage elasticities around 0.3. This is higher than previously estimated for physicians, in particular for those who are not self-employed.

Wage policy in the health care sector: a panel data analysis of nurses' labour supply.

Shortage of nurses is a problem in several countries. It is an unsettled question whether increasing wages constitute a viable policy for extracting more labour supply from nurses. In this paper we use a unique matched panel data set of Norwegian nurses covering the period 1993-1998 to estimate wage elasticities. The data set includes detailed information on 19,638 individuals over 6 years totalling 69,122 observations. The estimated wage elasticity after controlling for individual heterogeneity, sample selection and instrumenting for possible endogeneity is 0.21. Individual and institutional features are statistically significant and important for working hours. Contractual arrangements as represented by shift work are also important for hours of work, and omitting information about this common phenomenon will underestimate the wage effect.

A low-key social insurance reform--effects of multidisciplinary outpatient treatment for back pain patients in Norway.

This paper estimates treatment effects for back pain patients using observational data from a low-key social insurance reform in Norway. Using a latent variable model, we estimate the average treatment effect (ATE), the average effect of treatment on the treated (TT), and the distribution of treatment effects for multidisciplinary outpatient treatment at three different locations. To estimate these treatment effects, we use a discrete-choice model with unobservables generated by a factor structure model. Distance to the nearest hospital (in kilometres) is used as an instrument in estimating the different treatment effects. We find a positive effect of treatment of around 6 percentage points on the probability of leaving the sickness benefits scheme after allowing for selection effects and full heterogeneity in treatment effects. We also find that there are sound arguments for expanding the multidisciplinary outpatient programme for treating back pain patients.

Keeping nurses at work: a duration analysis.

A shortage of nurses is currently a problem in several countries, and an important question is therefore how one can increase the supply of nursing labour. In this paper, we focus on the issue of nurses leaving the public health sector by utilising a unique data set containing information on both the supply and demand side of the market. To describe the exit rate from the health sector we apply a semi-parametric hazard rate model. In the estimations, we correct for unobserved heterogeneity by both a parametric (Gamma) and a non-parametric approach. We find that both wages and working conditions have an impact on nurses' decision to quit. Furthermore, failing to correct for the fact that nurses' income partly consists of compensation for inconvenient working hours results in a considerable downward bias of the wage effect.