Factors influencing upward referral system of patients in Nairobi County

INTRODUCTION

The referral system forms main health systems coordinating a mechanism that ensures the harmonious movement of patients between different levels of health care institutions for effective and efficient service delivery. It refers to a situation where health facilities in the lower level without the technical-know-how and equipment required for treatment advice patients to seek for treatment from higher health facilities with better medical technology and technical skills.^1,2 Patient movement along the referral system is a day-to-day process within the health care setting and is an essential aspect in service delivery within the health sector. There are two types of medical referrals: Horizontal (referrals among health facilities of same level and vertical (referrals among health facilities of different levels). The healthcare system in Kenya has six levels as stipulated in the Kenya health policy 2014-2030 which are segmented into four tiers: tier I-community, tier II-dispensaries & health centres, tier III-County hospitals and tier IV-National referrals.³ There are numerous organizational factors associated with causing bottle necks in referral system across the 4 tiers of healthcare system. not much is known about this therefore the study was able to venture more on these contributing factors to enable policy makers reach informed decision. An effective referral system in any health system for service provision was aimed at facilitating harmonious liaison for the healthcare levels and patients getting required treatment at their convenience.⁴ Healthy referrals should be to enable patients get the needed healthcare at affordable prices, instill cost-effectiveness in all its components and timely dispensing of treatments.² However, this is not the case in the health referral system in Nairobi County. Kenyatta national hospital (KNH) operates beyond capacity around the year despite there being other referral hospitals in the county. KNH has a bed capacity of 1,800 beds with the average daily number of out-patients ranging from 2,500 and 3,000. This implies a bed utilization rate of over 170%, which sometimes can stretch to 300%.⁵ Many patients who seek treatment at Kenyatta National Hospital could be attended to tier III or even lower level health facilities. The high number of patients cause delay for patients suffering from complicated health problems, which ideally require specialized care. The study sought establishes the factors influencing upward referral system of patients for health service delivery in Nairobi County. Specific objectives were to establish the influence of knowledge of referral system, complexity of the disease, equipment required, and proximity to the referral hospital on upward referral system of patients in Nairobi County.

METHODS

This study adopted descriptive-cross sectional research design. The study site was Kenyatta National hospital, in Nairobi Kenya. The study sample was 239 respondents who were health workers in tier 2 and 3 health facilities in Nairobi County. Quantitative data was collected using a structured questionnaire which was self-administered. The questionnaire was pre-tested to ascertain validity and reliability of the tool. Data was analyzed using SPSS version 25 for descriptive and inferential statistics. Correlation and regression analysis were employed to test the influence of independent variables on the dependent variable. The regression model to measure the predictive power of the independent variables on the dependent variable was formulated as:

Where; Y=Upward movement for service delivery; X1=Knowledge of the referral system, X₂=complexity of the disease; X₃=equipment required; X₄=proximity to the referral hospital; β1_, β_2, β₃ and β₄=coefficients of determination and ε=error term. Research permit was obtained from the National commission for science, technology and innovation, Kenya, Nairobi County health office, and management of health facilities. Informed consent was sought from all respondents.

RESULTS

A total of 239 questionnaires were distributed but only 204 (85%) were fully filled and complete and included in this analysis. Social-demographic characteristics are presented in (Table 1).

Findings show that nearly half 84 (41%) of the respondents were from level 5 health facilities. Most respondents 150 (74%) were nurses by profession, 130 (64%) were aged 30-39 years, 116 (57%) were female and 119 (58%) had a diploma level of education. About half 99 (49%) had worked in their current facilities for 3-5 years and 96 (47%) had more than 10 years’ work experience. The respondents were asked to indicate whether they were aware of the existence of Kenya health sector referral guidelines. Results show that all the respondents were aware of the existence of Kenya health sector referral guidelines. However, just slightly over half of the respondents 109(53%) of the respondents said they always followed the referral guidelines set while referring a patient to a higher health facility level. The study further sought the respondent’s knowledge of the upward referral system. The results are shown in (Table 2).

SD=Strongly disagree, D=Disagree, N=Neutral, A=Agree, SA= Strongly agree

Results show that majority of the respondents disagreed that they were fully aware of the content in the referral strategy (mean=3.2892), that the referral strategic document is always available for reference (mean 3.0833), that they always have a copy of the referral strategy in the department (mean 3.0392), that the referral guidelines were easy to understand (mean 2.8431), that they always have continuous medical education on efficient referral systems (mean=2.4637). The findings imply that the healthcare workers were not knowledgeable of the referral procedures as stipulated by the referral guidelines and strategies. The study asked the participants to indicate their degree of agreement to statements on upward referral system (Table 3). Results show that the respondents disagreed that they often had to decide if to refer or not which was a delicate balance between quality and safety of healthcare (mean=3.2745) and that they always had to consult their peers before referring a patient (mean=3.1127), that they always refer patients in order to share responsibility on patient care (mean=2.7598) and that they sometimes hesitate to refer patients because of fear of being branded incompetent by colleagues or supervisors (mean=2.5294). In addition, the respondents disagreed that they often refer patients in order to conform to institutional expectations (mean=2.2255), that they often refer patients in order to have time to take care of other issues (mean=2.0343) and that they hesitate to refer because they always find writing a referral note to be tedious (mean=1.9853).

Moreover, the respondents strongly disagreed that they always encouraged to refer patients because of fear of cost of treatment (mean=1.8186) and that they were always encouraged to refer patients because their insurance does not cover treatment costs at our facility (mean=1.6078). This implies that although the respondents were not aware of the referral guidelines as shown in (Table 2), they understand the ethical issues behind the referral system. The study determined reasons why healthcare workers did not follow the set referral guidelines when referring a patient to a higher health facility level. See (Table 4). Results show that majority 173 (85%) of the respondents indicated inadequacy of time to follow set guidelines as a reason for not following the set referral guidelines. Further, 66 (32%) indicated that they were not sure about what the guidelines stipulates and that the disease was very severe and was a life-threatening condition as indicated by 114 (56%). Moreover, 113 (55%) of the respondents indicated that patient insisted to be referred and nearly half 114 (56%) said there was poorly coordinated referral system in the facility. The respondents were asked whether there was any inappropriate referrals that have happened in their health facility. Nearly all 186 (91%) of the respondents indicated that there were unnecessary referrals within their facilities, more than half 116 (57%) said that there was poor quality of referral documentation and 140 (69%) indicated that there was lack of communication during the referral process.

SD=Strongly disagree, D=Disagree, N=Neutral, A=Agree, SA= Strongly agree

On complexity of patient disease during referral, the respondents agreed that they often referred a patient when they felt a case was beyond their capacity to handle (mean=3.9167) and that most patients they referred were of clear-cut situations (mean=3.4245) (Table 5). The respondents however, disagreed that they are often faced with complex cases which needed immediate referral (mean=3.3775) and that they often referred patients with chronic disease (mean=2.9265). Results on the influence of equipment required for specialized treatment on upward referral are shown in (Table 6). Results showed that the respondents agreed that there were sophisticated diagnostic technologies at the referral hospitals (mean=3.7451) and that there were more advanced therapeutic technologies at the referral hospital (mean=3.6814). However, the respondents disagreed that the hospital were well equipped with critical care beds and ventilators (mean=3.2794); and that they have necessary equipment within their health facility (mean=3.0000). The influence of proximity to the referral hospital by the referring health workers showed that most agreed they always consider distance to the referral facility in order to reduce cost of utilization of health services by the patient (mean=3.8627), and that costs of transportation to reach health facilities affects referral of patients (mean=3.7745). The respondents however disagreed that patient’s decline referral due to the distance to the referral hospital (mean=3.3039). The study sought to establish the relationships that existed between the study variables. The results on bivariate analysis are shown in (Table 7). Findings on knowledge of referral system show a positive and significant correlation with upward referral system (r=179^*, p=0.011) In addition, findings on complexity of patient diseases how a positive and significant correlation with upward referral system (r=097, p=0.170). However, equipment required and proximity to the referral hospital had a positive relationship to upward referral system but it was not statistically significant. A general linear model was used to determining the projective influence of the four independent variables in the upward referral system of patients. Coefficient of determination (R²) gives an interpretation of the degree of change in the dependent variable explained by the changes in the independent variables or the variation in percentage of the dependent variable (upward referral system); as interpreted by the study’s independent variables (proximity to the referral hospital, equipment required, complexity of patient disease and knowledge of referral system).

*Correlation is significant at the 0.05 level (2-tailed). **Correlation is significant at the 0.01 level (2-tailed).

Results on the model summary revealed that the four independent variables in the study can explain about 82% of factors that influence upward referral system in Nairobi County. Further ANOVA results showed that the model was statistically significant in the prediction of the magnitude and direction of the influence that proximity to the referral hospital, equipment required, complexity of patient disease and knowledge of referral system had upward referral system in Nairobi County (p<0.0001) and F=14.922) at 5% significance level. A multivariate regression analysis was done to establish the degree to which independent variables in a combined relationship influenced upward referral system in Nairobi County. The results are shown in (Table 8). In this study the regression equation is: Y = β0 + β1X1 + β2X2 + β3X3 + β4X4 + ε

Becomes:

The constant (β₀) is also positive and significant (p< 0.001), which indicates that upward referral system will always exist at a certain minimum even without the one factor (proximity to the referral hospital) under investigation in this study. The coefficient of X₄ that is (β₄=0.640, p<0.002), indicates that a unit increase in the proximity to the referral hospital index leads to an improvement of the upward referral system. However, the other factors (knowledge of referral system, complexity of patient disease, and equipment required), under investigation in this study revealed a positive significant but had no statistical significant upward referral systems in a combined relationship.

DISCUSSION

The results showed that need for advanced diagnostic and therapeutic equipment influenced the need for upward referral in Nairobi County. The findings are agrees with that the need for diagnostic tests influenced the upward refferal of patients in the Malaysian referral system⁶. In addition the findings are rejected the findings bythat screening, diagnosis and assessment of respiratory diseases did not influence the upward reffearal of patients in Catalonian health referral system.⁷ Further the findings rejected the findings of a study bythat life support equipments did not influence the upward movement of patients in the referral system of Brazil.⁸ Finally, the findings are in agreeement with the findings of a stsudy by that a significant positive connection exited between the support equipment and upward movement of patients in the referral system of the United Kingdom.⁹

In this study proximity of the referral hospital influenced upwards referral system. The findings of the study are contrary to the findings by that the geographical distance to the referral hospital insignificantly influenced the upward movement of patients in the Ugandan referral system.¹⁰ However, they agree with findings by that the distance from primary healthcare centers vis-a-vis the distance from referral hospitals significantly influenced upward movement of patients in the referral system.¹¹

The findings on knowledge on referral to influence the upward movement of patients in the referral system have also been corroborated.¹² Further, findings by found that knowledge of the transfer system had positive significant influence on upward movement of patients in the referral system in the rural societies.¹³ The findings on complexity of the patient disease in this study disagree with a study that found chronic health conditions played a significant role in the movement of patients from care providers to referral hospitals in the USA.¹⁴ In addition, the findings are contrary to the findings by chronic health conditions is statistically significant in determining the upward movement of patients within the Iowa's referral system.

On the influence of equipment on upward referral systems, the findings agreed with the findings by that screening, diagnosis and assessment of respiratory diseases did not have any significant influence on the upward movement of patients in Catalonian health referral system.¹⁵ In addition the study findings agreed with the findings that life support equipment had a negative influence on the upward movement of patients in the referral system of Brazil.¹⁶ Also the study findings did not agree that there was a positive and significant influence of the life support equipment and upward movement of patients in the referral system of the UK.⁹ Further the study findings are contrary to the findings that the geographical distance to the referral hospital insignificantly influenced the upward movement of patients in the Ugandan referral system.¹⁰

Also the study findings agreed with the findings a study that the distance from primary healthcare centers vis-a-vis the distance from referral hospitals significantly influenced upward movement of patients in the referral system.¹¹ The study findings agrees with that provision of health care services is done at different levels and is determined by the kind of healthcare intervention that is needed by the patient.¹⁷

Findings agree with that referral can be explained as the process where health service provider at lower tier hospital for the lack of capabilities, resourceful facilities or both in the management of a specific clinical issues, request for help from providers who are equipped better or with specialized training to take them through the management or the treatment or to take over the treatment process as a whole.¹¹ Moreover the study findings agrees with the study that concluded that the improvement of quality and the raised rates of referrals is likely to promote the rates of utilizing the health care services.¹⁸

Limitations

The study had limitations that it focused only on public health facilities within Nairobi County. For this reason, the findings should be generalized with caution.

CONCLUSION

The study concludes that the healthcare workers were not fully aware of the content in the referral strategy and that the referral strategic document was not always available for reference by the health workers. The respondents often referred patients when they felt a case was beyond their capacity to handle, most patients they referred were of clear-cut medical cases and they referred patients because they did not have the latest treatment guidelines. The study concludes that there were sophisticated diagnostic technologies at the referral hospitals which also prompted the workers to refer patients. Proximity to the referral health facility played an important role in upward referral systems in Nairobi County. In summary, it therefore follows that, this study found statistical and significant evidence that proximity to the referral hospital in a combined relationship, significantly influences the upward referral systems in Nairobi County.

Recommendations

The study recommends that the Nairobi County Health Department should organize professional medical education to equip heath workers with update knowledge on an efficient referral system, the health facility management should ensure that each department in its facility has a copy of the referral policy and the Nairobi County Health Department should improve the communication systems between the lower levels health facilities and the higher or specialized facilities to ensure there a smooth referral system in Nairobi County

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: The study was approved by the Institutional Ethics Committee

REFERENCES

1. 1. Barnum H, Kutzin J. Public hospitals in developing countries.Available at: https://documents.worldbank.org/en/publication/documents-reports/documentdetail/919871468740383421/public-hospitals-in-developing-countries-resource-use-cost-financing. Accessed on 20 June 2021.
2. 2. Josefien V, Criel B. Studies in health services organization and policy. Available at: https://www.itg.be/Files/docs/shsop/shsop27.pdf. Accessed on 20 June 2021.
3. 3. Kenya Health Policy 2014-2030. http://publications.universalhealth2030.org/uploads/kenya_health_policy_2014_to_2030.pdf. Accessed on 20 June 2021.
4. 4. Besley T, Gouvela M. Alternative systems of health care provision. Health Policy J. 2014;9(19):53-8.
5. 5. Dutta A, Maina T, Ginivan M, Koseki S. Kenya health financing system assessment time to pick the best path: Palladium. Available at: http://www.healthpolicyplus.com/pubs.cfm?get=11323. Accessed on 20 June 2021.
6. 6. Foster J. The influence diagonistic tests on the upward movement of patients in the Malaysian Referral System: Malaysia healthcare. Available at: https://apps.who.int. Accessed on 20 June 2021.
7. 7. Legorreta AP, Schaff SR, Leibowitz AN, van Meijgaard J. Measuring the Effects of Screening Programs in Asymptomatic Employees: Detection of Hypertension Through Worksite Screenings. J Occup Environ Med. 2015;57(6):682-6.
8. 8. Bethune K. Influence of life support equipment on the upward movement of patients in the referral system of Brazilian hospitals: Available at: https://core.ac.uk/download/pdf/236410009.pdf. Accessed on 20 June 2021.
9. 9. Smith SA. Influence of life support equipment on the upward movement of patients in the referral system of the UK. Available at: https://apps.who.int/iris/bitstream/handle/10665/254716/9789290225164-eng.pdf. Accessed on 20 June 2021.
10. 10. Malgo A. Influence of geographic distance to the hospital on the upward movement of patients in Ugandan referral referral system. Available at: https://www.who.int/management/facility/RapidAssessmentofReferralCareSystems.pdf. Accessed on 20 June 2021.
11. 11. Devesh, Rosenbaum. Why is distance important for hospital choice? separating home bias from transport costs. J Indus Eco. 2021:15:14-9.
12. 12. Oluseye OM. Knowledge and Utilization of referral system among health care workers in selected primary health care centres in Oyo State, Nigeria. J Commu Med Prim Health Care. 2017;31(1):67-75.
13. 13. Elliot LC. The society for vascular surgery practice guidelines on the care of patients with an abdominal aortic aneurysm. J Vascular Surg. 2018;67(1):277.
14. 14. De Regge M, De Pourcq K, Meijboom B, Trybou J, Mortier E, Eeckloo K. The role of hospitals in bridging the care continuum: a systematic review of coordination of care and follow-up for adults with chronic conditions. BMC Health Serv Res. 2017; 17(1):550.
15. 15. Velickovski F. Influence of screening, diagnosis and assessment of respiratory diseases on the Upward movement of patients in Catalonian Health Referral system. Available at: https://www.euro.who.int/__data/assets/pdf_file/0010/98398/wa540ga199heeng.pdf. Accessed on 20 June 2021.
16. 16. Bethune K. Influence of life support equipment on the upward movement of patients in the referral system of Brazilian Hospitals. Available at: https://niti.gov.in/planningcommission.gov.in/docs/reports/genrep/rep_uhc0812.pdf. Accessed on 20 June 2021.
17. 17. Singh S, Doyle P, Campbell OM, Murthy GVS. Management and referral for high-risk conditions and complications during the antenatal period: knowledge, practice and attitude survey of providers in rural public healthcare in two states of India. Reprod Health. 2019;16(1):1-14.
18. 18. Preker AS, Carrin, G. Health financing for poor people resource mobilization and risk sharing. Available at: https://www.scirp.org. Accessed on 20 June 2021.