The purpose of this study was to examine changes in patient health indicator measurements in Finnish primary healthcare by comparing traditional care in 2017 with team-based care in 2022.
The study was a quantitative retrospective longitudinal cohort study using register data from patients' electronic health records, including low-density lipoprotein (LDL) cholesterol, haemoglobin A1c (HbA1c) levels, and blood pressure measurements. In this study, we compared patients in two groups: the traditional care group and the team-based care group. These groups were compared at measurement periods: first period of 2–6 months, second period of >6–10 months, and third period of >10–14 months.
Statistically significant differences were noted between the team-based care and traditional care groups over time in LDL (p = 0.005) and HbA1c (p < 0.0001). Although team-based care showed a significant change from baseline in systolic blood pressure during the second period, no between-group differences were observed over time.
Limited studies have been published to comparing traditional care with team-based care, both internationally and in Finland. Our study showed that team-based care may support the management of patients whose clinical monitoring includes LDL and HbA1c values and therefore have positive impacts on treatment outcomes.
Introduction
In traditional primary care, the responsibility for patient care lies with a single healthcare professional (Bodenheimer and Smith, 2013). Traditional healthcare practice is often structured so that the patient is primarily served by either a physician (medical doctor) or a registered nurse (RN) (Mukiapini et al., 2018). During the physician's or nurse's appointment, the patient describes their symptoms and needs, and the professional assesses the situation. This physician or nurse will carry out examinations and treatments and give advice and instructions. In this traditional care, care practices can be very protocol-based, and treatment decisions are often made based on a discussion between the patient and nurse or the patient and physician. Collaboration between healthcare professionals may be limited or it may occur in a way which the patient sees each professional separately without interprofessional collaboration. This kind of traditional care is weak in terms of collaboration and information flow between different professionals, which can lead to duplication or breaks in the care chain (Teperi et al., 2010). In today's world, where the emphasis is on a more comprehensive and interprofessional approach, there is a need for more collaborative and interprofessional approaches to care. This has led an increasing number of healthcare providers to seek new ways of organising their practice. One possibility is to move towards team-based care, which challenges the traditional way of thinking. (Rosen et al., 2019)
The increasing prevalence of chronic diseases in the present day has placed growing demands on healthcare systems, for which traditional care is no longer sufficient (Li et al., 2024). The integration of health technologies, along with the large amount of information now accessible to both professionals and patients, has made it increasingly challenging to manage complex patient comorbidities (Mukiapini et al., 2018; Bodenheimer et al., 2014). Physician shortages and an ageing population with multiple chronic conditions challenge the sustainability of traditional care. Time constraints also prevent physicians from delivering all necessary care by themselves. Delegating aspects of chronic disease management and preventive care to other healthcare professionals could help reduce physicians' workload and improve access to services. (Leach et al., 2017) In addition, the work culture in outpatient care in primary healthcare has shifted, and nurses' responsibilities have expanded. Nurses play a central role in managing chronic conditions and conduct a large share of patient visits in acute care settings. (Jolanki and Sinervo, 2016)
The definition of team-based care has been widely adopted in several healthcare studies. Team-based care means that two or more health professionals work together with patients and their caregivers to deliver health services – according to each patient's preferences – to achieve shared goals across care settings and provide coordinated, high-quality care. (Baik, 2017; Mitchell et al., 2012) Characteristics of team-based care are interprofessional collaboration, a patient-centred approach and integrated care processes. Prerequisites for achieving team-based care include collaboration between two or more professionals from different disciplines, e.g. a nurse or physiotherapist, an understanding of each team member's roles and responsibilities, a team climate characterised by mutual respect and adequate organisational support. The outcomes of team-based care are linked to the patient, the health professionals and the health organisation. (Baik, 2017)
Many different kinds of team-based care models exist such as the Patient-Centered Medical Home (PCMH) (Coniglio, 2013), Disease Management Programs (DMP) (DePasquale and Fonarow, 2014), Chronic Disease Management Programs (CDMP) (Gbigbi-Jackson et al., 2024), Multidisciplinary Team (MDT) (Taberna et al., 2020) and Teamlet model (Ghorob and Bodenheimer, 2015). These models were all created to improve patient outcomes and enhance the efficiency of care delivery.
Team-based care has been associated with numerous positive outcomes for patients as well as for healthcare professionals and organisations. It reduces medication utilisation, mortality, morbidity, hospitalisations, readmissions and healthcare expenditures, while improving patients' self-care, empowerment, engagement in preventive health behaviours, satisfaction and quality of life. (Szafran et al., 2019; Baik, 2017; Proia et al., 2014) When patients receive care from well-organised teams, care coordination becomes easier and patients access professionals more quickly – improving operational efficiency and reducing waiting times (Ansell et al., 2017; Wagner et al., 2017). It also enhances continuity of care, because a smaller cohesive team can better manage and maintain a comprehensive understanding of the patient's situation (Jokelin et al., 2025). For healthcare professionals, it enhances job satisfaction, productivity and retention, alongside reductions in burnout and intent to leave. Finally, at the organisational level, team-based care supports efficient allocation and use of resources, enhances access to services, strengthens care coordination and ultimately leads to improved economic outcomes (Baik, 2017). Healthcare costs have decreased due to shorter hospital stays as well as reduced hospital admissions, emergency visits, repeat treatment episodes and rehospitalisations (Matthys et al., 2017; Kravetz and Walsh, 2016).
Internationally, team-based care is recognised as a key strategy for reforming primary healthcare. It plays a central role in managing chronic diseases, caring for ageing populations and integrating mental health services into primary care. By combining complementary expertise, team members can optimise time management, balance workloads and provide more comprehensive, patient-centred care. (Li et al., 2024; Liu et al., 2023) Together, these findings highlight the multidimensional benefits of team-based care within healthcare systems. Despite these findings, key challenges in team-based care include ensuring effective communication between team members, managing different levels of expertise and roles, coordinating care across disciplines and sustaining a patient-centred approach. These challenges can be associated with limited resources, time constraints and sometimes resistance to change in healthcare (Tapani et al., n.d.). There are limited studies comparing traditional care with team-based care, both internationally and in Finland.
The purpose of this study was to examine changes in patient health indicator measurements in Finnish primary healthcare by comparing traditional care in 2017 with team-based care in 2022.
The research question was:
What changes, if any, were identified in patients' health indicators over time between those who received traditional care and those who received team-based care?
Materials and methods
Design and data collection
The study was conducted as a quantitative retrospective longitudinal cohort study using register data from patients' electronic health records, including low-density lipoprotein (LDL) cholesterol, haemoglobin A1c (HbA1c) levels and blood pressure measurements. These health indicators were chosen by the organisation as part of treatment effectiveness criteria. In this study, patients in two groups were compared: the traditional care group and the team-based care group. The patients in the groups were selected based on the chosen timeframe from one primary care centre in one well-being services county in Finland. The organisation introduced a team-based working model in the autumn of 2018, with the orientation phase already initiated in the spring of 2018. Consequently, the baseline data were drawn from 2017, a period during which the traditional primary healthcare service model was still operational. The year 2022 was selected for comparison under the assumption that team-based care had been fully established and integrated into practice by that time.
Selection criteria for each group
Data from the traditional care group (2017) and team-based care group (2022) were analysed using baseline measurements collected between 1 January and 30 April 2017. The comparison between traditional and team-based care was conducted solely based on measurements collected between 1 May 2017 and 30 June 2018 for the traditional care group and between 1 May 2022 and 30 June 2023 for the team-based care group. Patients were followed for 14 months after baseline, with inclusion requiring at least one measurement obtained two months after baseline. The data collection timeframe was divided into measurement periods: baseline, first period of 2–6 months, second period of >6–10 months and third period of >10–14 months. If a patient had multiple measurements within these time periods, the latest available measurement was selected.
Health indicators
Finnish primary healthcare treats a large number of patients with chronic diseases, such as cardiovascular disease, hypertension and diabetes. To assess the management and control of these conditions, three routinely collected clinical indicators were selected as study variables: low-density lipoprotein (LDL cholesterol), blood pressure, and HbA1c. These indicators are widely used in clinical practice and treatment guidelines to monitor disease status, evaluate treatment effectiveness and assess the risk of adverse health outcomes. For example, for reducing cardiovascular disease risk, effective management of elevated blood pressure (McEvoy et al., 2024) and LDL cholesterol levels (Stanciulescu et al., 2023) is essential, while HbA1c serves as a key indicator of long-term glycaemic control in the management of diabetes mellitus (Eyth et al., 2025). In this study, commonly recommended target levels were used as clinical reference values: LDL cholesterol <3.0 mmol/L, blood pressure <140/90 mmHg and HbA1c < 48 mmol/mol; however, individual treatment targets may vary.
Statistical analysis
Descriptive statistics were presented as means with standard deviations (SD) for all health indicators except HbA1c, which was presented as the median with lower and upper quartiles (Q1, Q3). Differences between the team-based care and traditional care groups over time were assessed using a linear mixed model for repeated measures including group, time and the interaction between them. Results from the model were presented as means with standard errors (SE) and mean changes over time within each care group (within-group changes) and differences in these changes between the team-based care and traditional care groups (between-group differences) with 95% confidence intervals (95% CI). HbA1c was log-transformed for the analyses, and the results were back-transformed to the original scale. Dunnett's adjustment was applied to pairwise comparisons of follow-up periods with baseline for all outcomes. Over the years, it is possible that some participants received both traditional and team-based care. In our data, having measurements in both groups was uncommon. Therefore, modelling within-participant correlation using participant-level random effects was considered sufficient to account for resulting non-independence of observations. All analyses were performed in R (version 4.1.1) with a two-sided significance level of 0.05. The performance package was used to assess model diagnostics.
Ethics
The research was carried out in compliance with internationally recognised standards of good scientific practice and ethical principles relevant to health sciences (ALLEA, 2023). Ethical approval was obtained from the local Ethics Committee for Human Sciences (decision 29 January 2024). In addition, authorisation to use the patients' electronic health records was granted by the organisation (decision SATAHA dn:o 2024-950). Because the data were collected from existing records, individual patient consent was not required. The data were processed in a closed, secure environment, with access only to designated researchers. As the data did not contain patient background information, the patients are not identifiable.
Results
This study examined changes in patient health indicators associated with traditional care and team-based care. The dataset consisted of 540 patients. The data of LDL measurements were obtained from 250 patients, systolic and diastolic blood pressure from 326 patients and HbA1c measurements from 318 patients. In baseline measurements, the patients' mean (SD) LDL as well as systolic and diastolic blood pressure levels were below the reference thresholds of 3.0 mmol/l and 140/90 mmHg, respectively, in both groups, while the median (Q1, Q3) HbA1c levels were higher in both groups. At baseline, team-based and traditional care differed in LDL (p = 0.033) and systolic blood pressure (p = 0.0004). As shown, LDL levels were significantly higher in the team-based group compared with the traditional group, while systolic blood pressure was lower in the team-based care group. (Table 1)
Baseline comparison between traditional care and team-based care
| Health indicator | N | na | nb | Traditional care | Team-based care | p-value |
|---|---|---|---|---|---|---|
| LDL | 250 | 130 | 137 | 2.53 (0.95) | 2.87 (1.19) | 0.033 |
| Systolic blood pressure | 326 | 113 | 230 | 139.3 (16.65) | 133.5 (15.66) | 0.0004 |
| Diastolic blood pressure | 326 | 113 | 230 | 78.1 (10.00) | 79.7 (10.30) | 0.278 |
| HbA1c | 317 | 166 | 191 | 49.0 (43.0, 55.0) | 49.0 (41.0,56.5) | 0.663 |
Note(s): N = number of patients in the whole sample (some patients may belong into both groups due to the long duration of treatment)
na = number of patients in traditional primary care
nb = number of patients in team-based care
Health indicators are given as mean (SD), except HbA1c, which is given median (Q1, Q3)
p-values are from a linear mixed model for repeated measures
Changes in patients' health indicators over time compared between traditional and team-based care
Among patients in team-based care, the model-based mean (SE) LDL, systolic and diastolic blood pressure levels and HbA1c levels remained below the respective reference thresholds of 3.0 mmol/l, 140/90 mmHg and 48 mmol/mol across all three follow-up time periods, whereas in traditional care, only the LDL and diastolic blood pressure levels remained below the reference thresholds. There were statistically significant differences between the team-based care and traditional care groups over time in LDL (p = 0.005) and HbA1c (p < 0.0001). Although the team-based care group showed a significant change from baseline in systolic blood pressure during the second period, no between-group differences were observed over time. (Tables 2 and 3)
Model-based means followed by the standard error of the mean over time
| Health indicator | Time in months | N | na | nb | Traditional care | Team-based care | p-value (time x group) |
|---|---|---|---|---|---|---|---|
| LDL | Baseline | 250 | 130 | 137 | 2.61 (0.09) | 2.84 (0.08) | 0.005 |
| 2–6 | 54 | 20 | 34 | 2.48 (0.18) | 2.55 (0.14) | ||
| >6–10 | 79 | 34 | 46 | 2.65 (0.14) | 2.36 (0.13) | ||
| >10–14 | 178 | 100 | 89 | 2.53 (0.09) | 2.44 (0.10) | ||
| Systolic blood pressure | Baseline | 326 | 113 | 230 | 139.7 (1.44) | 133.6 (1.02) | 0.162 |
| 2–6 | 117 | 36 | 82 | 139.0 (2.36) | 130.4 (1.58) | ||
| >6–10 | 154 | 59 | 99 | 138.5 (1.87) | 130.0 (1.46) | ||
| >10–14 | 212 | 91 | 128 | 142.1 (1.57) | 131.2 (1.30) | ||
| Diastolic blood pressure | Baseline | 326 | 113 | 230 | 78.5 (0.91) | 79.7 (0.64) | 0.440 |
| 2–6 | 117 | 36 | 82 | 77.3 (1.49) | 77.8 (1.00) | ||
| >6–10 | 154 | 59 | 99 | 77.4 (1.19) | 78.4 (0.92) | ||
| >10–14 | 212 | 91 | 128 | 79.9 (0.99) | 78.9 (0.83) | ||
| HbA1c | Baseline | 317 | 166 | 191 | 49.4 (0.77) | 49.8 (0.74) | <0.0001 |
| 2–6 | 69 | 35 | 37 | 51.4 (1.29) | 47.0 (1.14) | ||
| >6–10 | 142 | 63 | 84 | 51.6 (1.05) | 47.9 (0.88) | ||
| >10–14 | 230 | 124 | 126 | 51.3 (0.85) | 47.5 (0.78) |
Note(s): N = number of patients in whole the sample (some patients may belong into both groups due to the long duration of treatment)
na = number of patients in traditional primary care
nb = number of patients in team-based care
Model-based within-group mean changes from baseline and between-group differences with 95% confidence intervals
| Within-group mean change from baseline (95% CI)a | Between-group differences (95% CI)b | ||||
|---|---|---|---|---|---|
| Health indicator | Time in months | Traditional care | Team-based care | Team-based care vs. traditional care | p-value |
| LDL | 2–6 | −0.12 (−0.54, 0.29) | −0.29 (−0.62, 0.03) | −0.17 (−0.70, 0.36) | 0.755 |
| >6–10 | 0.05 (−0.28, 0.37) | −0.48 (−0.77, −0.20) | −0.53 (−0.96, −0.10) | 0.011 | |
| >10–14 | −0.07 (−0.28, 0.14) | −0.40 (−0.62, −0.19) | −0.33 (−0.63, −0.03) | 0.025 | |
| Systolic blood pressure | 2–6 | −0.70 (−6.46, 5.06) | −3.26 (−7.09, 0.63) | −2.53 (−9.46, 4.41) | 0.700 |
| >6–10 | −1.21 (−5.92, 3.49) | −3.59 (−7.16, −0.01) | −2.37 (−8.28, 3.54) | 0.645 | |
| >10–14 | 2.39 (−1.62, 6.39) | −2.47 (−5.96, 0.75) | −4.86 (−9.99, 0.28) | 0.069 | |
| Diastolic blood pressure | 2–6 | −1.19 (−4.83, 2.45) | −1.92 (−4.36, 0.52) | −0.72 (−5.11, 3.66) | 0.937 |
| >6–10 | −1.14 (−4.11, 1.84) | −1.27 (−3.53, 0.99) | −0.14 (−3.87, 3.60) | 0.997 | |
| >10–14 | 1.42 (−1.11, 3.95) | −0.75 (−2.79, 1.28) | −2.18 (−5.42, 1.07) | 0.269 | |
| HbA1cc | 2–6 | 1.04 (0.99, 1.10) | 0.94 (0.90, 0.996) | 0.91 (0.84, 0.98) | 0.008 |
| >6–10 | 1.04 (1.001, 1.09) | 0.96 (0.93, 0.999) | 0.92 (0.87, 0.98) | 0.0019 | |
| >10–14 | 1.04 (1.01, 1.07) | 0.96 (0.93, 0.99) | 0.92 (0.88, 0.96) | <0.0001 | |
Note(s): Statistically significant values (p < 0.05) shown in italic
Model-based mean changes from baseline within each care group (95% CI)
Model-based differences in changes from baseline between the team-based care and traditional care groups (95% CI)
Geometric mean ratio (95% CI)
Among the three follow-up periods, LDL level in the team-based care group showed significant decreases from baseline during the second and third periods, whereas no significant changes were shown in the traditional care group. The decrease from baseline was larger in the team-based care group than in the traditional care group at the second (−0.53; 95% CI −0.96, −0.10) (p = 0.011) and third periods (−0.33; 95% CI −0.63, −0.03) (p = 0.025). However, the reductions were not large enough for LDL values to differ significantly between groups at the second (−0.29; 95% CI −0.66, 0.07) and third follow-up periods (−0.09; 95% CI −0.34, 0.15).
HbA1c increased in traditional care by approximately 4% from baseline at each follow-up period, whereas it decreased in team-based care by approximately 6% during the first period, and by 4% during the second and third periods. Compared to the traditional care group, the team-based care group showed a statistically significant decrease from baseline by approximately 9% (geometric mean ratio (GMR) 0.91; 95% CI 0.84, 0.98) (p = 0.008) during the first period, 8% (GMR 0.92; 95% CI 0.87, 0.98) (p = 0.0019) during the second period and 8% (GMR 0.92; 95% CI 0.88, 0.96) (p < 0.0001) during the third period.
Discussion
Our study suggests that team-based care may improve the management of patients, as indicated by LDL and HbA1c levels, compared to traditional care. According to the study, the LDL levels of patients receiving traditional care showed no significant changes from baseline, while in team-based care, there was a significant decrease during the second and third follow-up periods. The changes from baseline to the second and third periods differed significantly between groups, suggesting that team-based care showed greater reductions than traditional care. Baseline values may partly explain this finding: LDL levels in the team-based care group were significantly higher than in the traditional care group, which may have contributed to the larger reductions in team-based care.
Previous studies support our finding that team-based care improves patients' disease control (Ballard-Hernandez and Sall, 2023; Pany et al., 2021). A previous study highlighted similarities and differences among dyslipidaemia treatment guidelines and also recognised that team-based care might be the best approach to manage dyslipidaemia (Ballard-Hernandez and Sall, 2023). For hyperlipidaemia, LDL measurements showed that patients treated by teams achieved disease control more frequently than those treated by solo providers (Pany et al., 2021). In addition, patients receiving team-based care had lower LDL levels than those receiving usual care (Pany et al., 2021; Bozovich et al., 2000).
There were no significant differences between the team-based care and traditional care groups in changes over time in patients' systolic or diastolic blood pressure. However, according to previous studies, patients achieve better disease control when care is delivered through team-based care (Pany et al., 2021; Kravetz and Walsh, 2016). Since we did not have access to background information on the patients selected for our study, such as possible medication use related to blood pressure treatment, it is difficult to say why a corresponding result was not seen. Due to the lack of background information, there is also no information on how sick patients were in the study compared with those in previous studies. Pany et al. (2021) found that after receiving an initial abnormal biomarker result, diagnostic entries for hypertension were made significantly more often in teams than by solo providers. Previous research has also shown a significantly greater reduction in blood pressure in the team-based care group. After four months, systolic blood pressure had decreased statistically significantly, while diastolic blood pressure had also decreased, though not to a statistically significant degree. At the post-intervention measurement point, 62% of patients in team-based care had lower systolic blood pressure and 54% had lower diastolic blood pressure, compared to 41% and 37%, respectively, in the usual care group – both statistically significant differences (Kravetz and Walsh, 2016).
According to this study, patients' HbA1c levels showed a significant increase from baseline over the second and third follow-up periods in traditional care, whereas in team-based care, there was a significant decrease from baseline at each follow-up period. As a result, changes from baseline to each follow-up period differed significantly between the groups. Pany et al. (2021) found in their study that after receiving an initial abnormal biomarker result, diagnostic entries were made significantly more often by teams than by solo providers for type 2 diabetes. One study also compared the time it took to achieve an HbA1c level of less than 7% between patients managed by a pharmacist–physician team and those receiving usual medical care. The results showed that patients who collaborated with a pharmacist as part of team-based care reached their target HbA1c levels in a shorter amount of time (Cowart et al., 2020).
While improvements in LDL and HbA1c were observed, the study design does not allow us to attribute these changes solely to the presence of a multiprofessional team. However, a team-based primary healthcare model could potentially improve population health and equity, enhancing the quality and efficiency of nursing and medical practice through collaboration and could provide patients with more accessible, coordinated and personalized care, leading to better outcomes and satisfaction. Previous studies have shown that team-based primary care reduces emergency room visits and hospital stays. This saves public funds and shifts resources from expensive corrective care to cheaper prevention. (Rajaratnam et al., 2025)
Strengths and limitations
This study had some strengths. First, quantitative methods make it possible to efficiently analyse large datasets, such as registry data. The health indicators used in this study were utilised in the numerical measurement of specific phenomena (LDL, blood pressure and HbA1c) and these have been in use for a long time and have proven effective. Second, the data used were collected in real-world patient care. Third, before team-based care started, the employees were trained in team-based work, which could be expected to provide a better foundation for similar collaborative practices. Fourth, this study generates new knowledge while reinforcing the existing evidence, which is currently limited.
However, the study also had some limitations. First, internal validity and external generalisability may be limited by the data, which were originally collected for other purposes and may contain gaps or biases. Furthermore, generalisability is limited by the fact that the study was conducted in a single health and social services centre within one well-being services county in Finland. Second, differences in background characteristics (e.g. age, comorbidities, medication use and beginning of treatment) between patients measured at different time points and in different groups (traditional vs. team-based) could have biased our results. It should also be noted that some patients may have been included in both groups because the conditions are chronic and can coexist. This may have reduced the independence of the groups and should be considered when interpreting the results. Third, it should be noted that there were significant differences between traditional care and team-based care in the baseline LDL and systolic blood pressure measurements at the starting point, which might also influence these results. Fourth, the healthcare landscape changed between 2017 and 2022, for example through developments in digital health infrastructure, meaning that the observed findings may not be attributable solely to the implementation of team-based care.
Conclusion
Our study showed statistically significant changes in LDL and HbA1c values in team-based care. No significant differences were observed between the team-based and traditional care groups in patients' blood pressure. Team-based care appears to improve lipid and glycaemic control but the absence of an effect on blood pressure suggests its benefits may be outcome-specific, highlighting the need for further research to clarify its overall clinical impact. Team-based care utilises the expertise of different professionals, and a single professional is supported by the expertise of others. In this way, the patient receives the best and most versatile treatment possible. However, more research is needed on why the results for LDL and HbA1c seem to be better with team-based care. In the future, it would also be important to conduct comparative research on different operating models at the same time, for example by comparing the operating practices of different organisations. A mixed methods study could examine team-based healthcare by using quantitative data (e.g. patient outcomes, access, satisfaction scores) to measure effectiveness, alongside qualitative methods (e.g. interviews or focus groups) to explore how patients experience care, how teams function and what factors influence satisfaction and outcomes.
