Responsiveness of Inpatient Services at the Public General Hospitals in Gaza

This study aimed at measuring the responsiveness level of inpatient services according to patients’ perspectives, and identifying the factors affecting its performance at the Gazan public general hospitals, particularly in internal medicine and surgical departments, in 2020. It was a cross-sectional descriptive study, conducted in 5 public general hospitals in Gaza. Data were collected between October 2019 and June 2020 in a survey included 675 inpatients using an interviewed questionnaire. Data were analyzed using software SPSS 22.0 and by applying descriptive statistics, independent t-test and analysis of variance (ANOVA), and post hoc Scheffé test. The overall responsiveness was “above moderate”, 80.15% and 80.9%, for both internal medicine and surgical departments, respectively. Access to social support, condentiality and dignity were the highest-performing domains, while, choice of provider and quality of basic amenities were the lowest. Dignity and condentiality were the most important domains based on patients’ viewpoints. Hospital type, medical department, and hospital stay in addition to the educational level, marital employment status, income, and insurance were the most factors on the of responsiveness and its domains.


Background
Responsiveness is a key strategic feature and stands as one of the three intrinsic goals of any health system, hence, it can act as an effective tool in evaluating its performance 1,2 . Improvement of responsiveness plays a substantial role in enhancing the other two goals, including health status and nancial protection, and ultimately the health system performance 3 .
Differently from the quality of care, responsiveness represents the legitimate expectations of people in terms of non-clinical aspects of the health care services 4,5 . Therefore, health systems seek to ameliorate these aspects and respond to peoples' expectations in order to raise the level of satisfaction regardless of the impact of responsiveness on health status 1,6 . In addition, sustainment of responsiveness is not nancially costy when compared to interventions related to health promotion 7 .
Responsiveness is expressed through two dimensions; the level and distribution. The level of responsiveness can be measured based on the perception of users, while, the distribution shows to what extent services are fairly provided to individuals in the society 8 .
Furthermore, responsiveness is comprised of eight domains which can be classi ed into two categories: 1) interpersonal domains including; dignity (DIG), clear communication (CC), autonomy (AUT) and con dentiality (CON), and 2) the client-orientation domains encompassing; prompt attention (PA), access to social support networks (ASN), choice of provider (CP) and quality of basic amenities (QBA) 8,9 . In the World Health Report 2000, 191 countries have been ranked according to their estimated responsiveness (level and distribution) using questionnaires developed by the World Health Organization (WHO). Some countries in the Mediterranean Region were involved such as Egypt, Jordan, Syria, Lebanon, Iran and Qatar, and were ranked 102, 85, 70, 55, 100, and 27, respectively 8 . State of Palestine was not among the targeted countries in that study. The Palestinian health system embraces ve health care providers. The Ministry of Health (MOH) is the main provider which owns and administers 471 primary healthcare centers and 26 public hospitals in both West Bank and Gaza Strip (GS). Of the 26 hospitals, there are 13 in the GS containing about 2,240 active beds, in other words, 11.4 bed per 10,000 inhabitants. Out of 13, 7 hospitals in the GS are general as they provide internal medicine and surgical inpatient services. The active beds in internal medicine and surgical departments stand for 52% of the total active beds at the public hospitals in GS 10,11 .
Generally, there is no focus on the responsiveness as a core goal of the Palestinian health system. Since there is a paucity of studies in the literature on measuring the level of responsiveness for any types of health services at any level in Palestine, this study was the rst in the Palestinian health context which aimed at measuring the responsiveness level of inpatient services according to patients' perspectives, and identifying the factors affecting its performance at the Gazan public general hospitals, particularly in internal medicine and surgical departments in 2020.

Study setting
This cross-sectional descriptive survey was conducted in the internal medicine and surgical departments at the Gazan public general hospitals in 2020. Anonymously, the included hospitals in this study which locate in the ve governorates, were given the codes from A to E. Hospital A (large and non-educational), B (central and educational complex), C (large and non-educational), D (central and educational complex) and E (central and educational large hospital). It is noteworthy that all hospitals in our study deliver a similar range of services, and this can signi cantly reduce the case-mix bias and make the comparison more feasible.

Sampling and sample size
Multi-staged sampling has been used to select the study participants. Firstly and based on the annual admissions in the investigated departments, the total sample size was calculated for both internal medicine and surgical departments by applying the formula below, 381 and 382 participants, respectively. n=N*X / (X + N -1) Where, X = Z α/2 2 *p*(1-p) / MOE 2 , and Z α/2 is the critical value of the Normal distribution at α/2 (e.g. for a con dence level of 95%, α is 0.05 and the critical value is 1.96), MOE is the margin of error, p is the sample proportion, and N is the population size 12 . N in the current study was 39,000 and 49,000 admissions in internal medicine and surgical departments, respectively.
Secondly, proportional sampling was performed to determine the sample in each hospital in accordance to the number of installed beds in the abovementioned departments. Next, another proportional sampling was carried out according to the number of beds in male and female wards at the same department. Finally, participants were selected randomly after con rming the inclusion criteria; a) inpatients admitted in the departments under study, b) patients whose age ≥ 16 year old, and c) patients who have stayed in the department for 2 days or more. On the other hand, patients hospitalized in intensive care units, cardiology care units, and obstetrics departments have been excluded.

Data collection
By modifying the WHO questionnaire, a pre-tested interview-based questionnaire was developed. The English version has been translated into Arabic by two professional persons, then back-translation from Arabic to English has been undertaken once again by another two professionals. Content validity, addition and elimination of items to modify the original questionnaire have been performed by 12 experts in the eld from academia and health sector. A pilot study of 150 patients has been conducted to check the reliability of the modi ed questionnaire which covers two main parts; 1) the patients' sociodemographic characteristics including: age, sex, marital status, educational level, occupational status, income, insurance, hospital stay, in addition to the hospital of admission and medical department, and 2) the perception of patients toward the eight domains that encompass 37 items on a 4-point Likert scale (1 = highly disagree, 2 = disagree, 3 = agree, and 4 = highly agree). The idea of adopting even-numbered response categories is to have a balanced number of positive and negative options as well as prevent giving people an "out" and makes them more thoughtful in their responses 13

Data analysis
Subsequent to assessing data normality by the two-sample Kolmogorov-Smirnov test, data were analyzed by applying descriptive statistics (frequency, percentage, mean, and standard deviation), independent t-test and analysis of variance (ANOVA) using SPSS 22.0 (SPSS Inc., Chicago, IL, USA). A Scheffé post-hoc test was also employed to compare between groups' means in ANOVA. For all tests, results was regarded statistically signi cant when a p value ≤ 0.05. After calculating the average for each domain score, the weighted mean score per each was then determined as recommended in the WHO methodology 8 . The percentage scores (%) used in Tamimi study have been adopted as references for evaluating responsiveness and its domains; (20-36% very low), (36.1-52% low), (52.1-68% moderate), (68.1-84% above moderate) and (84.1-100% high) 14 .

Ethical considerations
Ethical approvals for this study were granted by the Vice-Chancellor in Research Affairs-Tehran University of Medical Sciences (IR. TUMS. VCR. REC. 1398.360), in addition, the study was submitted to the ethics committee local to our study in Palestine, Helsinki committee in the Palestinian Health Research Council, approved and given the code: (PHRC/HC/959/19). Moreover, informed consents were verbally obtained out from all participants, and con dentiality was assured as well.

Participant characteristics
As illustrated in Table 1  Scores of the responsiveness domains Table 2 shows the ndings for both overall responsiveness and the eight domains of inpatient services in the departments under study. The overall responsiveness was "above moderate", 80.15% and 80.9%, at internal medicine and surgical departments, respectively. ASN was the highest in its performance level (98%), and ranked as "high" based on the study reference values. This was followed by CON, DIG and CC based on participants' opinions in both internal medicine and surgical departments, whilst, CP and QBA were evaluated lowest in both departments and their levels ranged from 58 to 62% (moderate). Drawing on patients' viewpoints in both departments, DIG and CON were the most important domains, whereas, QBA and CP were reported 6th and 8th, respectively. of stay have led to statistically signi cant differences in the level of total responsiveness as well as some domains (Table 3). Scheffé test has investigated the statistically signi cant effect of certain groups on responsiveness ( Table 4). Hospital of admission was the most in uential factor as it revealed statistically signi cant differences in responsiveness among the investigated hospitals ( Table 5).
The scores of QBA have demonstrated remarkable variations with respect to different groups within the examined variables, followed by PA, CC and AUT, ASN and CP, and at last, DIG and CON (Table 4).    DIG, CON and PA were the most important domains in both departments, while, CP and QBA were the least. This was, to some extent, in line with some studies in the literature 16,25,26 . This might be attributed to the importance of respect, privacy and receiving rapid and timely care when needed, on the other hand, people realize that they have no chance to choose their providers based on the policy of the MOH, accordingly, they underestimate the importance of CP. CC and ASN were the seventh and fourth important domains for patients in internal departments, but the third and seventh in surgical departments, respectively. Generally, working environment and the procedures employed may differ from one department to another, in addition, these results necessitate the focus on some domains more than others.
Furthermore, ASN, CON, DIG, CC, AUT and PA were the highest-performing domains in both departments, respectively, whilst, CP and QBA were the least. Well-scheduled visits for escorts and relatives, and the value of social support in the view of married persons might justify the highest level of ASN as stated in two previous studies 16,17 . For CP, generally, the Palestinian health system does not allow patients to choose their providers, rather, patients should receive the health services in speci c places according to the geographical distribution and place of residency due to the shortage of specialists and the potentially consequent bias. The best performing item in AUT domain was that related to the informed consent prior to any medical or surgical procedure (about 90%), and this can be due to the rules and regulations which mandate lling out certain forms accordingly. This corresponds the ndings involved in other studies 13,17,22 . In CON domain, the privacy of information and medical records was the most paramount for patients in both departments, and this was consistent with some former studies 18,27,28 . It should be noted that the siege and related nancial hardship and economic decline, staff burnout, shortages of in healthcare workforce, and the problem in remuneration system may adversely affect the staff commitment, and ultimately reduced responsiveness performance, particularly PA domain 27,29 .
Type of hospital was one of the most in uential factors on domains' performance. Except for QBA, hospital E was the best-performing among all hospitals based on patients' perspectives, however, hospital D was better in its performance in QBA. E is a large, teaching and referral hospital which was well-established, administered and operated by European staff for years. Moreover, well-quali ed specialists, rules and regulations, management practices, sophisticated electronic system, educational programs, and less crowding play a crucial role in raising awareness of health care providers about non-clinical aspects of care. The information system in such hospitals produces rich data and leads to enhancing the provider-client relationship 20,30 . The recent construction and maintenance activities, and the newly founded and well-facilitated buildings have contributed in ameliorating QBA in departments under study in hospital D. Conversely, lack of rooms' space, electricity cutoffs in the absence of alternatives such as diesel, have decreased level of QBA at the Nigerian public hospitals 20 . In their study, Mosallam et al discovered that rooms shared by several patients without curtains have adversely affected the CON performance at the public hospitals.
Elderly patients in this study (≥ 70 years old) showed higher levels of responsiveness, CC, AUT, PA and QBA, and this was consistent with Peltzer et al study 21 . These ndings could be due to the cultural respect to elderly people which gives them the advantage in all domains. Low expectations and insu cient awareness level, in addition to the information asymmetry between providers and loweducated patients might account for the better performance in responsiveness, CC, PA and QBA. In the same vein, jobless patients and those of low income gave higher performance ratings for PA and QBA, and AUT, PA, CP and QBA, respectively. This also re ects the low ceiling of expectations, the lower social status they may feel, and the conditions they live as well. However, these results could give a positive impression but might mask some defects in the system as stated by Mohammed et al 30 .
Although there was notable differences in the levels of DIG, CON, ASN and QBA according to the insurance type, this might be referred to reasons other than the type itself that is because the Palestinian health system does not limit the accessibility of uninsured persons to inpatient services. For that, there is a way to enable those uninsured patients and those holders of expired insurance to receive the health care service and sign a commitment to pay for the insurance later on. Nonetheless, the expectations of enrollees in the labor or social insurance plans regarding QBA may justify their better perspective about this domain.
Patients admitted to surgical departments reported higher performance for AUT and QBA, and these differences were statistically signi cant. The clear information and consultation on treatment alternatives in surgical departments, moreover, the quality of basic services related to rooms and facilities in these departments may owe for the better-off performance therein. The ndings in the present study were consistent with those of previous studies 21, 24,30 . For hospital stay, it was obvious that those patients who have stayed for 6 days or more showed higher performance levels in CC, AUT, PA, CP and QBA. The friendly atmosphere which can be developed with time between those patients and the staff, in addition to the paternalistic behavior 30 are potential explanations for this result.
CP was selected as the least important domain according to patients from both departments. Furthermore, CP and QBA were the lowest domains in performance (about 60%). CP performance was signi cantly affected by hospital type (hospital E), income (≤ 1000 NIS) and hospital stay (≥ 6 days). However, there is a restricted room in hospital E for the patient to choose his provider as this hospital is relatively less crowded and has plenty of well-quali ed specialists. Giving more attention to CP is expected to raise performance of the overall responsiveness 13,18,20,27 .

Conclusion
The assessment of the responsiveness of inpatient health care services at public hospitals is useful in integrating the health system goals. The overall responsiveness was about 80%, above moderate, and is viable for further improvement. The domains QBA, CP and PA were highlighted as crucial areas to ameliorate the perceived responsiveness of healthcare services. The gap between importance and performance of CC and ASN in internal medical and surgical departments, respectively, should be addressed. Reform strategies should be directed toward all domains, considering their weight and effect in promoting the total responsiveness. Attention should be also given to patients' as well as hospital characteristics in order to elucidate a comprehensive picture about responsiveness. Further researches of mixed-method design are required to obtain more rigorous results. Extensive studies are also needed to cover hospitals other than the public ones.

Declarations
Ethics approval and consent to participate: Ethical approvals for this study were granted by the Vice-Chancellor in Research Affairs-Tehran University of Medical Sciences (IR. TUMS. VCR. REC. 1398.360), in addition, the study was submitted to the ethics committee local to our study in Palestine, Helsinki committee in the Palestinian Health Research Council, approved and given the code: (PHRC/HC/959/19). Moreover, informed consents were verbally obtained out from all participants, and con dentiality was assured as well.
Consent for publication: Not applicable.
Availability of data and materials: The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.