4. Modelling Survival Data in Medical Research

Background
Little is known about the effectiveness of strategies to enable people to achieve and maintain recommended levels of physical activity.

Objectives
To assess the effectiveness of interventions designed to promote physical activity in adults aged 16 years and older, not living in an institution.

Search methods
We searched The Cochrane Library (issue 1 2005), MEDLINE, EMBASE, CINAHL, PsycLIT, BIDS ISI, SPORTDISCUS, SIGLE, SCISEARCH (from earliest dates available to December 2004). Reference lists of relevant articles were checked. No language restrictions were applied.

Selection criteria
Randomised controlled trials that compared different interventions to encourage sedentary adults not living in an institution to become physically active. Studies required a minimum of six months follow up from the start of the intervention to the collection of final data and either used an intention-to-treat analysis or, failing that, had no more than 20% loss to follow up.

Data collection and analysis
At least two reviewers independently assessed each study quality and extracted data. Study authors were contacted for additional information where necessary. Standardised mean differences and 95% confidence intervals were calculated for continuous measures of self-reported physical activity and cardio-respiratory fitness. For studies with dichotomous outcomes, odds ratios and 95% confidence intervals were calculated.

Main results
The effect of interventions on self-reported physical activity (19 studies; 7598 participants) was positive and moderate (pooled SMD random effects model 0.28 95% CI 0.15 to 0.41) as was the effect of interventions (11 studies; 2195 participants) on cardio-respiratory fitness (pooled SMD random effects model 0.52 95% CI 0.14 to 0.90). There was significant heterogeneity in the reported effects as well as heterogeneity in characteristics of the interventions. The heterogeneity in reported effects was reduced in higher quality studies, when physical activity was self-directed with some professional guidance and when there was on-going professional support.

Authors' conclusions
Our review suggests that physical activity interventions have a moderate effect on self-reported physical activity, on achieving a predetermined level of physical activity and cardio-respiratory fitness. Due to the clinical and statistical heterogeneity of the studies, only limited conclusions can be drawn about the effectiveness of individual components of the interventions. Future studies should provide greater detail of the components of interventions.

/pdf/interventions-for-promoting-physical-activity-461chrroe1.pdf

Interventions for promoting physical activity

This paper systematically reviews the scientific literature on the effects of individual and work-related factors on the Work Ability Index (WAI). Studies on work ability published from 1985 to 2006 were identified through a structured search in PubMed, and Web of Science. Studies were included if the WAI was used as measure of work ability and if quantitative information was presented on determinants of work ability. In total, 20 studies were included with 14 cross-sectional studies and six longitudinal studies. Factors associated with poor work ability, as defined by WAI, were lack of leisure-time vigorous physical activity, poor musculoskeletal capacity, older age, obesity, high mental work demands, lack of autonomy, poor physical work environment, and high physical work load. The WAI is associated with individual characteristics, lifestyle, demands at work, and physical condition. This multifactorial nature of work ability should be taken into account in health promotion programmes aimed at maintaining and promoting the participation of the labour force and improvement of the performance at work.

https://oem.bmj.com/content/oemed/66/4/211.full.pdf

The effects of work-related and individual factors on the work ability index: A systematic review

Telemedicine (TM) is the use of telecommunication systems to deliver health care at a distance. It has the potential to improve patient health outcomes, access to health care and reduce healthcare costs. As TM applications continue to evolve it is important to understand the impact TM might have on patients, healthcare professionals and the organisation of care.To assess the effectiveness, acceptability and costs of interactive TM as an alternative to, or in addition to, usual care (i.e. face-to-face care, or telephone consultation).We searched the Effective Practice and Organisation of Care (EPOC) Group's specialised register, CENTRAL, MEDLINE, EMBASE, five other databases and two trials registers to June 2013, together with reference checking, citation searching, handsearching and contact with study authors to identify additional studies.We considered randomised controlled trials of interactive TM that involved direct patient-provider interaction and was delivered in addition to, or substituting for, usual care compared with usual care alone, to participants with any clinical condition. We excluded telephone only interventions and wholly automatic self-management TM interventions.For each condition, we pooled outcome data that were sufficiently homogenous using fixed effect meta-analysis. We reported risk ratios (RR) and 95% confidence intervals (CI) for dichotomous outcomes, and mean differences (MD) for continuous outcomes.We included 93 eligible trials (N = 22,047 participants), which evaluated the effectiveness of interactive TM delivered in addition to (32% of studies), as an alternative to (57% of studies), or partly substituted for usual care (11%) as compared to usual care alone.The included studies recruited patients with the following clinical conditions: cardiovascular disease (36), diabetes (21), respiratory conditions (9), mental health or substance abuse conditions (7), conditions requiring a specialist consultation (6), co morbidities (3), urogenital conditions (3), neurological injuries and conditions (2), gastrointestinal conditions (2), neonatal conditions requiring specialist care (2), solid organ transplantation (1), and cancer (1).Telemedicine provided remote monitoring (55 studies), or real-time video-conferencing (38 studies), which was used either alone or in combination. The main TM function varied depending on clinical condition, but fell typically into one of the following six categories, with some overlap: i) monitoring of a chronic condition to detect early signs of deterioration and prompt treatment and advice, (41); ii) provision of treatment or rehabilitation (12), for example the delivery of cognitive behavioural therapy, or incontinence training; iii) education and advice for self-management (23), for example nurses delivering education to patients with diabetes or providing support to parents of very low birth weight infants or to patients with home parenteral nutrition; iv) specialist consultations for diagnosis and treatment decisions (8), v) real-time assessment of clinical status, for example post-operative assessment after minor operation or follow-up after solid organ transplantation (8) vi), screening, for angina (1).The type of data transmitted by the patient, the frequency of data transfer, (e.g. telephone, e-mail, SMS) and frequency of interactions between patient and healthcare provider varied across studies, as did the type of healthcare provider/s and healthcare system involved in delivering the intervention.We found no difference between groups for all-cause mortality for patients with heart failure (16 studies; N = 5239; RR:0.89, 95% CI 0.76 to 1.03, P = 0.12; I(2) = 44%) (moderate to high certainty of evidence) at a median of six months follow-up. Admissions to hospital (11 studies; N = 4529) ranged from a decrease of 64% to an increase of 60% at median eight months follow-up (moderate certainty of evidence). We found some evidence of improved quality of life (five studies; N = 482; MD:-4.39, 95% CI -7.94 to -0.83; P < 0.02; I(2) = 0%) (moderate certainty of evidence) for those allocated to TM as compared with usual care at a median three months follow-up. In studies recruiting participants with diabetes (16 studies; N = 2768) we found lower glycated haemoglobin (HbA1c %) levels in those allocated to TM than in controls (MD -0.31, 95% CI -0.37 to -0.24; P < 0.00001; I(2)= 42%, P = 0.04) (high certainty of evidence) at a median of nine months follow-up. We found some evidence for a decrease in LDL (four studies, N = 1692; MD -12.45, 95% CI -14.23 to -10.68; P < 0.00001; I(2 =) 0%) (moderate certainty of evidence), and blood pressure (four studies, N = 1770: MD: SBP:-4.33, 95% CI -5.30 to -3.35, P < 0.00001; I(2) = 17%; DBP: -2.75 95% CI -3.28 to -2.22, P < 0.00001; I(2) = 45% (moderate certainty evidence), in TM as compared with usual care.Seven studies that recruited participants with different mental health and substance abuse problems, reported no differences in the effect of therapy delivered over video-conferencing, as compared to face-to-face delivery. Findings from the other studies were inconsistent; there was some evidence that monitoring via TM improved blood pressure control in participants with hypertension, and a few studies reported improved symptom scores for those with a respiratory condition. Studies recruiting participants requiring mental health services and those requiring specialist consultation for a dermatological condition reported no differences between groups.The findings in our review indicate that the use of TM in the management of heart failure appears to lead to similar health outcomes as face-to-face or telephone delivery of care; there is evidence that TM can improve the control of blood glucose in those with diabetes. The cost to a health service, and acceptability by patients and healthcare professionals, is not clear due to limited data reported for these outcomes. The effectiveness of TM may depend on a number of different factors, including those related to the study population e.g. the severity of the condition and the disease trajectory of the participants, the function of the intervention e.g., if it is used for monitoring a chronic condition, or to provide access to diagnostic services, as well as the healthcare provider and healthcare system involved in delivering the intervention.

/pdf/interactive-telemedicine-effects-on-professional-practice-1xxbw8tg81.pdf

Interactive telemedicine: effects on professional practice and health care outcomes

Extensive information is available from official statistics and descriptive studies on the association between different socio-demographic background factors and sickness absence. This information addresses age, gender, place of residence, and socio-economic status. However, few studies have thoroughly analysed these background factors, and rigorous scientific evidence on the causal relationship between these factors and sick leave is lacking. Regarding the family, we found no scientific evidence that marital status or children living at home were associated with sickness absence. However, we found limited scientific evidence for an effect of divorce. Regarding work-related factors, we found limited scientific evidence for an effect of physically stressful work, and moderate scientific evidence for low psychological control over the work situation. We found limited scientific evidence for a correlation in time between unemployment and sickness absence, but insufficient scientific evidence for the causes of the association. There was moderate scientific evidence that the amount of sickness absence is influenced by the design of the social insurance system, but insufficient evidence on the magnitude of change required to influence the level of sickness absence. Essentially the same results apply to disability pension, although the number of studies is small. However, we found moderate scientific evidence for the effects of socio-economic status, which could be explained partly by childhood experiences.

https://journals.sagepub.com/doi/pdf/10.1080/14034950410021853

Chapter 5. Risk factors for sick leave - general studies

Objectives This multicentered randomized controlled trial evaluated the effect of worksite exercise intervention on perceived work ability and sick leaves.

Methods Women (N=260, mean age 40 years) engaged in physically demanding laundry work were individually randomized into an intervention (N=133) or control (N=127) group. Perceived work ability was assessed with questionnaires at 3, 8, 12, and 15 months. Sick leave information was obtained from the personnel administration. Follow-up attendance was 100% at 3 months but declined gradually to 90% by 15 months. Both the intervention and control subjects received a 30-minute feedback on their physical capacity from a physiotherapist and individual exercise prescription and counseling. The intervention subjects also participated in worksite exercise training guided by a physiotherapist. Sixty-minute sessions (N=26) were held once a week for 8 months. About 50% of the intervention group participated in at least two-thirds of the sessions.

Results According to a dichotomized work ability index, at 12 months, workers with "good" or "excellent" work ability increased more in the intervention group than in the control group (11.0%, 95% CI 0.2-21.9), as did the health-related prognosis of work ability at 8 months (8.1%, 95% CI 0.5-16.3). There were no statistically significant differences between the two groups as regards job satisfaction, work ability index (including series of questions on 7 items), or sick leaves.

Conclusions Physical activity once a week at worksites improves the perceived work ability of women with physically demanding work only slightly. Perceived work ability and sick leaves cannot be affected very positively using single-component exercise intervention. Work ability promotion may need a more multiprofessional approach.

Effectiveness of a worksite exercise program with respect to perceived work ability and sick leaves among women with physical work.

Objectives 
To study the associations between self-reported health problems and sickness absence from work.

 Methods 
The results of a questionnaire survey were combined with archival data of sickness absence of 1341 employees (88% males; 62% blue-collar) in the construction, service and maintenance work within one corporation in Finland. Sex, age, and occupational grading were controlled as confounders. Zero-inflated negative binomial (ZINB) regression model was used in the statistical analysis of sickness absence data.

 Results 
The prevalence of self-reported health problems increased with age, from 23% in the 18-30 year old to 54% in the 55-61 year old. However, in the age group 18 to 30 years, 71% had been absent from work and in the age group 55 to 61 years this proportion was 53%. When health problems and occupational grading were accounted for in the ZINB model, age as such was not associated with the number of days on sick leave, but the young workers still had higher propensity for (any) sickness absence than the old. Self-rated future working ability and musculoskeletal impairment were strong determinants of sickness absence. Among those susceptible to have sick leaves, the estimated mean number of absence days increased by 14% for each rise of 1 unit of the impairment score (scale 0 to 10).

 Conclusions 
Young subjects had surprisingly high probability for sickness absence although they reported better health than their older colleagues. Higher total count of absence days was found among subjects reporting health problems and poorer working ability, regardless of age, sex and occupational grade. These findings have implications both for the management and health care system in the prevention of work disability.

https://oem.bmj.com/content/oemed/64/11/739.full.pdf

Self‐reported health problems and sickness absence in different age groups predominantly engaged in physical work

Objectives: To evaluate the effectiveness of two occupational health intervention programmes, both compared with usual care. Methods: Based on a health survey, 1341 employees (88% males) in construction, service and maintenance work were classified into three groups: ‘‘low risk’’ (n = 386), ‘‘intermediate risk’’ (n = 537) and ‘‘high risk’’ (n = 418) of sickness absence. Two separate randomised trials were performed in the groups ‘‘high risk’’ and ‘‘intermediate risk’’, respectively. Those high risk subjects that were allocated to the intervention group (n = 209) were invited to occupational health service for a consultation. The intervention included, if appropriate, a referral to specialist treatment. Among the intermediate risk employees those in the intervention group (n = 268) were invited to call a phone advice centre. In both trials the control group received usual occupational health care. The primary outcome was sickness absence during a 12month follow-up (register data). Results: The high risk group, representing 31% of the cohort, accounted for 62% of sickness absence days. In the trial for the high risk group the mean sickness absence was 30 days in the usual care group and 19 days in the intervention group; the mean difference was 11 days (95% CI 1 to 20 days). In the trial for the intermediate risk group the mean sickness absence was 7 days in both arms (95% CI of the mean difference –2.3 to 2.4 days). Conclusions: The identification of high risk of work disability was successful. The occupational health intervention was effective in controlling work loss to a degree that is likely to be economically advantageous within the high risk group. The phone advice intervention for the intermediate risk group was not effective in controlling work loss primarily due to poor adherence. Sickness absence may result in considerable personal and public financial consequences. Long-term sickness absence also predicts early retirement. 1–3

https://oem.bmj.com/content/oemed/65/4/236.full.pdf

The effectiveness of two occupational health intervention programmes in reducing sickness absence among employees at risk. Two randomised controlled trials.

Objectives: To determine whether, from a healthcare perspective, a specific occupational health intervention is cost effective in reducing sickness absence when compared with usual care in occupational health in workers with high risk of sickness absence. Methods: Economic evaluation alongside a randomised controlled trial. 418 workers with high risk of sickness absence from one corporation were randomised to intervention (n = 209) or to usual care (n = 209). The subjects in the intervention group were invited to occupational health service for a consultation. The intervention included, if appropriate, a referral to specialist treatment. Register data of sickness absence were available for 384 subjects and questionnaire data on healthcare costs from 272 subjects. Missing direct total cost data were imputed using a two-part regression model. Primary outcome measures were sickness absence days and direct healthcare costs up to 12 months after randomisation. Cost effectiveness (CE) was expressed as an incremental CE ratio, CE plane and CE acceptability curve with both available direct total cost data and missing total cost data imputed. Results: After one year, the mean of sickness absence was 30 days in the usual care group (n = 192) and 11 days less (95% CI 1 to 20 days) in the intervention group (n = 192). Among the employees with available cost data, the mean days of sickness absence were 22 and 24, and the mean total cost J974 and J1049 in the intervention group (n = 134) and in the usual care group (n = 138), respectively. The intervention turned out to be dominant—both cost saving and more effective than usual occupational health care. The saving was J43 per sickness absence day avoided with available direct total cost data, and J17 with missing total cost data imputed. Conclusions: One year follow-up data show that occupational health intervention for workers with high risk of sickness absence is a cost effective use of healthcare resources.

https://oem.bmj.com/content/oemed/65/4/242.full.pdf

An occupational health intervention programme for workers at high risk for sickness absence. Cost effectiveness analysis based on a randomised controlled trial.

Lack of physical activity (PA) is a known risk factor for many health conditions. The workplace is a setting often used to promote activity and health. We investigated the effectiveness of an intervention on PA and productivity-related outcomes in an occupational setting. We conducted a randomized controlled trial of 12 months duration with two 1:1 allocated parallel groups of insurance company employees. Eligibility criteria included permanent employment and absence of any condition that risked the participant’s health during PA. Subjects in the intervention group monitored their daily PA with an accelerometer, set goals, had access to an online service to help them track their activity levels, and received counseling via telephone or web messages for 12 months. The control group received the results of a fitness test and an information leaflet on PA at the beginning of the study. The intervention’s aim was to increase PA, improve work productivity, and decrease sickness absence. Primary outcomes were PA (measured as MET minutes per week), work productivity (quantity and quality of work; QQ index), and sickness absence (SA) days at 12 months. Participants were assigned to groups using block randomization with a computer-generated scheme. The study was not blinded. There were 544 randomized participants, of which 521 were included in the analysis (64% female, mean age 43 years). At 12 months, there was no significant difference in physical activity levels between the intervention group (n = 264) and the control group (n = 257). The adjusted mean difference was −206 MET min/week [95% Bayesian credible interval −540 to 128; negative values favor control group]. There was also no significant difference in the QQ index (−0.5 [−4.4 to 3.3]) or SA days (0.0 [−1.2 to 0.9]). Of secondary outcomes, body weight (0.5 kg [0.0 to 1.0]) and percentage of body fat (0.6% [0.2% to 1.1%]) were slightly higher in the intervention group. An exploratory subgroup analysis revealed no subgroups in which the intervention affected physical activity. No adverse events were reported. The intervention was not found effective, and this study does not provide support for the effectiveness of the workplace PA intervention used here. ClinicalTrials.gov identifier: NCT00994565

/pdf/the-effectiveness-of-physical-activity-monitoring-and-1frms3hvm4.pdf

The effectiveness of physical activity monitoring and distance counseling in an occupational setting - Results from a randomized controlled trial (CoAct)

Timo Aro

Papers

Effectiveness of a worksite exercise program with respect to perceived work ability and sick leaves among women with physical work.

Self‐reported health problems and sickness absence in different age groups predominantly engaged in physical work

The effectiveness of two occupational health intervention programmes in reducing sickness absence among employees at risk. Two randomised controlled trials.

An occupational health intervention programme for workers at high risk for sickness absence. Cost effectiveness analysis based on a randomised controlled trial.

The effectiveness of physical activity monitoring and distance counseling in an occupational setting - Results from a randomized controlled trial (CoAct)