ICD-Based Diagnoses Versus Admission Reason- Based Diagnostic Groupings in Out-Of-Hospital Intensive Care: Implications for NursingRelated Health Services Research

Secondary diagnoses were highly prevalent among patients receiving out-of-hospital intensive care. Overall, 71.7% of patients (n = 1,309) had a percutaneous endoscopic gastrostomy (PEG) tube. Common comorbidities included hypertension (21.9%, n = 399), epilepsy (21.8%, n = 398), heart disease (18.1%, n = 331), lung disease (13.6%, n = 249) and diabetes mellitus (12.3%, n = 224). Gastrointestinal disorders were present in 12.2% (n = 222) and psychiatric illnesses affected 10.2% (n = 187). Congenital physical impairments were documented in 10.1% (n = 184) and spasticity in 9.0% (n = 164).

Other neurological or systemic comorbidities included nerve disorders (8.9%, n = 163), vascular disease (8.4%, n = 154), kidney disease (8.4%, n = 153), congenital mental impairment (8.3%, n = 151) and bone or skeletal disease (8.1%, n = 148). Pain disorders (7.7%, n = 140) and thyroid disorders (6.6%, n = 121) were also reported. A small proportion of patients (6.6%, n = 120) had no secondary diagnoses documented.

Additional less frequent conditions included mild spasticity (6.1%), insulin-dependent diabetes mellitus (5.9%), allergies and urinary tract disorders (5.6% each). Obesity (BMI >30) was documented in 4.9%. Other metabolic or endocrine disorders, substance use, ophthalmologic disease, malignant tumours and severe spasticity were each present in ≤3% of the cohort. Rare conditions (<1%) included biliary or pancreatic disease, stomas, COPD grades 1–2, hypotension and various unspecified or palliative tumours.

Overall, these data demonstrate a high burden of multimorbidity among patients receiving out-of-hospital intensive care with a predominance of PEG use, neurological conditions, cardiovascular and metabolic comorbidities.

Secondary Diagnoses by PAR and Care Setting (tables not shown)

Secondary diagnoses were prevalent across all admission groups (PAR1–PAR5), reflecting high multimorbidity and complex care needs in patients receiving out-of-hospital intensive care (n = 1825). The distribution and type of secondary diagnoses differed notably between admission groups and care settings.

PAR1: Brain Impairment (n = 940)

Patients with brain impairment exhibited a high prevalence of PEG tube insertion (81.7%, 95% CI 79.2–84.2). Neurological comorbidities were common, with epilepsy in 29.7% (95% CI 26.8–32.6) and spasticity in 11.9% (95% CI 9.8–14.0). Cardiovascular conditions included hypertension (28.1%, 95% CI 25.2–31.0) and heart disease (18.4%, 95% CI 15.9–20.9) while pulmonary and metabolic comorbidities affected 13.8% and 13.5% of patients. Gastrointestinal disorders (13.0%, 95% CI 10.8–15.1), psychiatric illness (9.9%, 95% CI 8.0–11.8) and vascular disease (9.3%, 95% CI 7.4–11.1) were also present. 

PAR2: Neuromuscular Diseases (n = 450)

In patients with neuromuscular diseases PEG tubes were inserted in 65.1% (95% CI 60.7–69.5) and congenital physical impairments affected 21.6% (95% CI 17.8–25.4). Epilepsy (17.6%, 95% CI 14.0–21.1) was a notable neurological secondary diagnosis while congenital mental impairment occurred in 14.2% (95% CI 11.0– 17.4). Pulmonary (13.6%), cardiovascular (heart disease 12.9%; hypertension 8.4%) and musculoskeletal comorbidities (bone/ skeletal disease 12.0%; nerve disorders 10.7%) were also present.

Notably, 9.6% of patients had no rec-order secondary diagnoses.

PAR3: Spinal Cord Injury (n = 88)

Among spinal cord injury patients, half (50.0%, 95% CI 39.6– 60.4) had a PEG tube. Spasticity was observed in 22.7% (95% CI 14.0–31.5) with mild and severe forms in 15.9% and 3.4%. Gastrointestinal disorders (20.5%) and lung disease (19.3%) were frequent. Neurological (epilepsy 12.5%), cardiovascular (heart disease 12.5%; hypertension 14.8%) and urinary tract comorbidities (14.8%) were also present, along with obesity, bone/skeletal, kidney and psychiatric disorders. Only 6.8% had no secondary diagnoses.

PAR4: Lung and Respiratory Diseases (n = 279)

Patients with lung diseases had PEG tubes in 57.7% (95% CI 51.9– 63.5). Cardiovascular comorbidities were frequent such as heart disease (26.2%) and hypertension (25.8%). Psychiatric illness affected 16.8%, diabetes mellitus 16.1% and kidney disease 16.1%. Gastrointestinal disorders were present in 13.6%, lung disease in 12.5% and nicotine abuse in 11.1%. Less common secondary diagnoses (<10%) included vascular and nerve disorders, thyroid disease, pain, obesity, epilepsy and congenital impairments.

PAR5: Heart Diseases (n = 39)

In the small cohort of patients with heart diseases, PEG insertion and cardiovascular comorbidities were common but data were limited due to sample size. Hypertension and heart disease were the primary secondary diagnoses, while metabolic, neurological and other comorbidities occurred less frequently.

Care Setting: Individual Home Care (IHC) vs Shared Living Facility (SLF)

Patients receiving individual home care (IHC, n = 822) and those in shared living facilities (SLF, n = 1003) exhibited differences in secondary diagnoses burden.

IHC patients showed higher prevalence of spasticity (unspecified: 61.9%) and neurological comorbidities, including epilepsy (23.5%) and congenital physical or mental impairments (16.3% and 12.7%). Cardiovascular, pulmonary and metabolic comorbidities were less frequent than in SLF patients.

SLF patients had higher overall multimorbidity with PEG tubes, cardiovascular diseases and metabolic disorders more common. This is consistent with their older mean age (61.9 vs 40.0 years in IHC) and the predominance of AG1 and AG4 diagnoses in SLF settings.

Overall, secondary diagnoses in out-of-hospital intensive care patients reveal high prevalence of PEG use, neurological disorders, spasticity, cardiovascular and metabolic comorbidities.

Discussion

This study provides comprehensive insights into the population of patients receiving out-of-hospital intensive care (OHIC) and highlights the marked heterogeneity of this group, as well as substantial structural differences between care settings. By combining clinically informed diagnostic grouping with detailed analysis of secondary diagnoses, the findings offer important implications for clinical practice, nursing care and health policy within the evolving German regulatory framework.

Principal Findings and Interpretation

A key finding is the predominance of neurological conditions as the primary reason for admission to out-of-hospital intensive care. More than half of the patients were classified within the brain impairment group (51.5%), followed by neuromuscular diseases (24.7%) and respiratory conditions (15.3%). Within the brain impairment group, vascular events - particularly ischemic stroke and intracerebral haemorrhage -  constituted the largest subgroup. This distribution is consistent with previous literature identifying acquired brain injury as a major driver of long-term intensive care dependency and prolonged need for ventilatory support [7,8].

Clear differences emerged between care settings. Patients with brain impairment and respiratory diseases were predominantly treated in shared living facilities (SLF), whereas those with neuromuscular diseases were more frequently managed in individual home care (IHC). These findings align with known disease trajectories. Neuromuscular disorders are often characterized by progressive physical decline with relatively preserved cognition, enabling home-based care, whereas acquired brain injuries frequently result in severe cognitive and functional impairment requiring more structured care environments [9,10].

Importantly, only 29% of the study population were invasively or non-invasively ventilated. This finding challenges the common perception of OHIC as a predominantly ventilation-driven care sector and suggests that a substantial proportion of patients require intensive care primarily due to neurological impairment, functional dependency and complex nursing needs rather than respiratory failure alone. This observation aligns with emerging evidence that the concept of chronic critical illness extends beyond mechanical ventilation and includes a broader spectrum of longterm care dependency [11,12].

Advantages of Admission Reason-Based Diagnostic Grouping

A key methodological strength of this study is the use of clinically defined admission groups rather than relying solely on ICDbased classifications. While ICD-coding is indispensable for epidemiological and administrative purposes, it primarily reflects discrete etiological diagnoses and may insufficiently capture functional impairment, which is a major determinant of long-term care needs [12–14].

By aggregating diagnoses into eight clinically reasonable categories such as brain impairment, neuromuscular diseases, spinal cord injury and chronic respiratory disease, the present approach integrates both underlying pathophysiology and functional consequences. This strategy is consistent with established approaches in health services research, including hierarchical condition categories and comorbidity groupings, which have been shown to enhance care settings, interpretability and reduce heterogeneity in large datasets [15–17].

This classification enables a more clinically relevant representation of care complexity by linking diagnostic categories with functional status, care requirements and resource utilization. For example, diverse aetiologies of brain impairment - such as traumatic brain injury, hypoxic injury and stroke - associated common long-term care needs, including dysphagia management, PEG dependence, and neurorehabilitation. Grouping these conditions therefore avoids artificial fragmentation of clinically comparable populations and facilitates reasonable comparisons across care settings [12,15,16].

Notably, a substantial discrepancy was observed in the identification of dysphagia depending on the data source. Accurate identification of dysphagia is of particular importance, as it directly influences key aspects of care, including PEG placement, aspiration risk and overall nursing workload. Dysphagia is a wellestablished predictor of complications and resource utilization in neurologically impaired populations [10,18,19]. While clinical assessment and supplementary data indicated that 85% of patients were affected, ICD-based analysis identified dysphagia in only 16% of cases. This gap underlines the limitations of administrative coding in capturing clinically relevant functional impairments and further supports the added value of admission reason-based classification approaches [12,13,15,20].

Underestimation in ICD-based datasets may therefore result in incomplete assessment of patient needs, inadequate staffing models and a systematic underrepresentation of care complexity in out-of-hospital intensive care populations.

Multimorbidity and Care Complexity

Another central finding is the high prevalence of multimorbidity across the cohort. The mean number of secondary diagnoses was substantial and significantly higher in patients with brain impairment compared to those with neuromuscular diseases. The high prevalence of PEG feeding, dysphagia and tracheostomy reflects the complexity of long-term intensive care populations and is consistent with previous studies on chronically critically ill patients [11,12].

Common comorbidities included arterial hypertension and epilepsy. The overall incidence of multimorbidity observed in this study supports prior findings that patients requiring long-term ventilation represent a highly complex population with extensive medical and nursing needs [21].

Importantly, multimorbidity differed significantly between care settings. Patients in SLF exhibited a markedly higher number of secondary diagnoses compared to those receiving IHC. This corresponds with the higher mean age in SLF and supports the concept that institutionalized care settings disproportionately serve older and more medically complex patients [2,21].

The finding that just 29% of the patients are ventilated further emphasizes that nursing work-load in OHIC is not primarily defined by ventilation management but by the complexity of multimorbidity, functional impairment and long-term dependency in the presence of a tracheostomy. This has important implications for training, staffing and competency frameworks, which should extend beyond respiratory care.

Age Distribution and Structural Care Patterns

The age distribution demonstrates a wide spectrum with a substantial proportion of both younger and older patients, although more than one-third were aged over 65 years. The pronounced age difference between care settings suggests structural selection mechanisms: Younger patients are more frequently managed at home, while older and more multimorbid individuals are more often treated in shared living facilities.

These differences likely reflect not only clinical characteristics but also social determinants, such as the availability of informal caregivers and family support. Previous research has high-lighted the critical role of social context and caregiver resources in enabling home-based long-term ventilation [1].

Care Pathways and System-Level Implications

More than half of the patients were transferred to out-ofhospital intensive care from neurological rehabilitation settings, highlighting the importance of this transition interface. This finding is consistent with prior studies showing that out-of-hospital ventilation often represents a continuation of care following acute illness and rehabilitation rather than a primary care setting [7,14].

The clustering of specific diagnostic groups within particular care settings suggests the presence of implicit allocation mechanisms. However, these appear to be only partially standardized and may be influenced by structural and historical factors rather than clear clinical criteria. This observation aligns with ongoing discussions about the need for improved care coordination and standardized pathways in long-term ventilation care [15].

Implications of Secondary Diagnoses for Nursing Practice in OHIC

The detailed analysis of secondary diagnoses provides clinically meaningful insights into the specific care demands faced by nurses in out-of-hospital intensive care and highlights the importance of diagnosis-informed care planning. Across all admission groups, the high prevalence of multimorbidity confirms that nursing care in OHIC extends far beyond the management of a single primary condition and instead requires continuous prioritization, coordination and adaptation to complex and interacting health needs.

Patients with brain impairment (PAR1), who represent the largest subgroup, exhibit a particularly demanding care profile characterized by high rates of PEG dependence, dysphagia and epilepsy. For nursing practice, this translates into a combination of intensive nutritional management, aspiration prevention, seizure monitoring, and strict infection control procedures. The coexistence of neurological deficits and cardiovascular comorbidities further increases the need for continuous clinical assessment and early detection of deterioration. These findings reinforce previous evidence that neurologically impaired patients require highly skilled nursing care with a strong focus on complication prevention and long-term functional management [12,18].

In contrast, patients with neuromuscular diseases (PAR2) present a different but equally complex care profile. Although multimorbidity is comparatively lower, the higher prevalence of congenital impairments and progressive physical decline requires long-term continuity of care, patient-specific adaptation of assistive technologies and a strong emphasis on preserving autonomy. For nurses, this implies a shift from acute complication management toward long-term disease trajectory management and psychosocial support which has been identified as a core component of care in this population [1,9].

Patients with spinal cord injury (PAR3) highlight the importance of managing secondary complications such as spasticity, gastrointestinal dysfunction, and urinary tract disorders. These conditions are closely linked to daily nursing interventions, including positioning, bowel and bladder management, and prevention of secondary complications. The high variability of comorbidities in this group underscores the need for individualized care plans and specialized nursing competencies.

In patients with chronic respiratory diseases (PAR4), the coexistence of cardiovascular, metabolic and psychiatric comorbidities reflects a multidimensional care burden. For nurses, this requires integrated management of respiratory support, monitoring of comorbid chronic diseases and attention to mental health aspects, which are often underrecognized but contribute significantly to overall care complexity and patient outcomes [11,21].

Importantly, the comparison between care settings reveals structurally different nursing demands. Patients in shared living facilities (SLF) exhibit higher overall multimorbidity, including increased prevalence of PEG use, cardiovascular disease and metabolic disorders. This indicates a higher level of clinical complexity and justifies increased staffing levels, advanced competencies and structured care processes in these settings. In contrast, patients in individual home care (IHC) more frequently present with neurological conditions such as spasticity and epilepsy, requiring specialized neurological nursing skills and close monitoring despite a lower overall burden of systemic comorbidity.

Overall these findings demonstrate that secondary diagnoses are not merely additional clinical information but are central determinants of nursing workload, required competencies and care organization in OHIC. Systematic consideration of comorbidity patterns -particularly when structured through clinically reasonable admission groups- can support more accurate staffing models, targeted training programs and improved allocation of resources. Moreover, they provide a foundation for developing standardized care pathways that reflect the actual complexity of patients rather than relying solely on primary diagnoses or administrative classifications.

Integration into the German Policy Context (IPReG)

The findings of this study are highly relevant in the context of the german “Intensivpflege- und Rehabilitationsstärkungsgesetz” (IPReG) [22], which fundamentally restructured OHIC provision since October 29^th 2020.

IPReG introduced a new legal entitlement to out-of-hospital intensive care strengthened quality requirements and mandated regular assessments to ensure appropriate and needs-based care. It also explicitly aims to improve care quality and to reduce inappropriate long-term ventilation and/or tracheal cannulation.

A central objective of the reform is to align care provision more closely and to eliminate structural inefficiencies and misallocation of resources. In this context, the present findings provide important empirical support for several policy goals:

1. The demonstrated heterogeneity of patient populations highlights the need for differentiated, indication-based care allocation. The clear clustering of diagnostic groups across care settings suggests that such allocation mechanisms already exist but are not yet standardized.
2. The high incidence of multimorbidity - particularly in SLF - supports the need for quality-assured care structures and regular reassessment. The policy requirement for evaluation of care appropriateness directly corresponds to the complexity observed in this study.
3. The underestimation of key functional impairments such as dysphagia in administrative data highlights a critical challenge for the implementation of IPReG: Accurate needs assessment. Since eligibility and care planning increasingly depend on structured evaluations, reliance on incomplete coding systems may undermine the intended improvements in care quality.

Finally, the findings underscore the importance of qualified nursing staff and specialized competencies. The observed variability in care complexity suggests that standardized staffing ratios alone may be insufficient without incorporating patient-specific clinical profiles.

Limitations and Strengths

Several limitations should be considered. First, some variables were derived from different data sources, which may limit comparability. Second, the cross-sectional design precludes causal inference. Third, variability in documentation practices may have influenced the reporting of secondary diagnoses.

Additionally, important factors such as functional status and social determinants were not available but are likely to play a crucial role in care allocation and outcomes.

Strengths include the large nationwide sample, physician validation of diagnostic categories, and systematic documentation of nursing-relevant parameters.

Conclusion

This study demonstrates that OHIC is characterized by a highly complex, predominantly neurologically driven patient population with substantial multimorbidity. Significant differences between care settings in terms of age, diagnostic profile and comorbidity patterns indicate underlying structural allocation mechanisms.

By linking clinical findings with nursing practice and policy context, the results underline the need for functionally informed classification systems, differentiated care models, and evidencebased staffing approaches. These aspects are worth considering when implementing IPReG, as they provide a clearer understanding of patients’ needs and the complexity of care.

References

1. Lindahl B, Sandman PO, Rasmussén BH (2005) On becoming dependent on home mechanical ventilation. J Adv Nurs 49: 33-42.
2. Kojima G (2015) Prevalence of Frailty in Nursing Homes: A Systematic Review and Meta-Analysis. J Am Med Dir Assoc 16: 940-5.
3. World Health Organization (2022) ICD-11: International Classification of Diseases (ICD) [Internet]. [cited 2026 Apr 15].
4. Busse R, Geissler A, Aaviksoo A, Cots F, Häkkinen U, et al. (2013) Diagnosis related groups in Europe: moving towards transparency, efficiency, and quality in hospitals? BMJ. 7:346:f3197.
5. O’Malley KJ, Cook KF, Price MD, Wildes KR, Hurdle JF, et al. (2005) Measuring diagnoses: ICD code accuracy. Health Serv Res 40:16201639.
6. Cichero JAY, Altman KW (2012) Definition, prevalence and burden of oropharyngeal dysphagia: a serious problem among older adults worldwide and the impact on prognosis and hospital resources. Nestle Nutr Inst Workshop Ser 72:1-11.
7. Windisch W, Dreher M, Geiseler J, Siemon K, Brambring J, et al. (2017) S2k-Leitlinie: Nichtinvasive und invasive Beatmung als Therapie der chronischen respiratorischen Insuffizienz- Revision 2017. Pneumologie. 71: 722-795.
8. Needham DM, Davidson J, Cohen H, Hopkins RO, Weinert C, et al. (2012) Improving long-term outcomes after discharge from intensive care unit: report from a stakeholders’ conference. Crit Care Med 40: 502-9.
9. Bourke SC, Tomlinson M, Williams TL, Bullock RE, Shaw PJ, et al. (2006) Effects of non-invasive ventilation on survival and quality of life in patients with amyotrophic lateral sclerosis: a randomised controlled trial. Lancet Neurol 5:140-7.
10. Turner-Stokes L (2004) The evidence for the cost-effectiveness of rehabilitation following acquired brain injury. Clin Med 4:10-2.
11. Carson SS (2012) Definitions and Epidemiology of the Chronically Critically Ill. Respir Care 57: 848-58.
12. Nelson JE, Cox CE, Hope AA, Carson SS (2010) Chronic Critical Illness. Am J Respir Crit Care Med 182:446-54.
13. Cassini A, Högberg LD, Plachouras D, Quattrocchi A, Hoxha A, et al. (2019) Attributable deaths and disability-adjusted life-years caused by infections with antibiotic-resistant bacteria in the EU and the European Economic Area in 2015: a population-level modelling analysis. Lancet Infect Dis 19:56-66.
14. Schönhofer B, Geiseler J, Dellweg D, Fuchs H, Moerer O, et al. (2019) [Pro-longed Weaning - S2k-Guideline Published by the German Respiratory Society]. Pneumologie.73:723-814.
15. Ergan B, Oczkowski S, Rochwerg B, Carlucci A, Chatwin M, et al. (2019) European Respiratory Society guidelines on long-term home non-invasive ventilation for management of COPD. Eur Respir J 54:1901003.
16. Pope GC, Kautter J, Ellis RP, Ash AS, Ayanian JZ, et al. (2004) Risk adjustment of Medicare capitation payments using the CMS-HCC model. Health Care Financ Rev 25:119-141.
17. Charlson ME, Pompei P, Ales KL, MacKenzie CR (1987) A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 40:373-83.
18. Martino R, Foley N, Bhogal S, Diamant N, Speechley M, et al. (2005) Dysphagia after stroke: incidence, diagnosis, and pulmonary complications. Stroke. 36:2756-63.
19. Attrill S, White S, Murray J, Hammond S, Doeltgen S (2018) Impact of oropharyngeal dysphagia on healthcare cost and length of stay in hospital: a systematic review. BMC Health Serv Res 18:594.
20. Wilkins T, Gillies RA, Thomas AM, Wagner PJ (2007) The prevalence of dysphagia in primary care patients: a HamesNet Research Network study. J Am Board Fam Med JABFM 20:144-50.
21. MacIntyre EJ, Asadi L, Mckim DA, Bagshaw SM (2016) Clinical Outcomes Associated with Home Mechanical Ventilation: A Systematic Review. Can Respir J 2016:6547180.
22. Bundesministerium für Gesundheit (BMG). Intensivpflege- und Rehabilitationsstärkungsgesetz (IPReG) [Internet]. 2020 [cited 2026 Apr 16].