Factors Influencing Support for National Health Insurance: Evidence from Qassim Region, Saudi Arabia

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Title- Factors Influencing Support for National Health Insurance: Evidence from Qassim Region, Saudi Arabia.

 

My specific and suggestions comments are listed below.

• The abstract should not contain any undefined abbreviations (e.g. WTP) or values and percentages. the authors should avoid mention the results, method, conclusion. Readers can distinguish between the different meanings.
• The introduction section should usually not be subdivided. Please delete the subsection title 1-1.
• Please highlight the reminder of the manuscript in the last part of the introduction.
• I find that the introduction is unfocused and overly long, lacking a clear research question or motivation. Further, authors reiterate known information about insurance scheme in Saudi Arabia without identifying gaps or clarifying how this study advances the field of health insurance.
• Page – 7. Results are overly descriptive and need to be interpreted with greater clarity. For example: How do you explain that participants living in urban areas are more likely to participate in the health insurance program as compared to those living in rural or suburban areas?
• Page – 12. Line 364. What is Health Belief Model (HBM) and what is the relationship with the results above?
• Subsections “Contributions, implications and limitations” should be integrated into the conclusion section.

Good luck with your paper.

Author Response

Sl.No	Comments	Responses
Reviewer-1
1	The abstract should not contain any undefined abbreviations (e.g. WTP) or values and percentages. The authors should avoid mention the results, method, conclusion. Readers can distinguish between the different meanings.	We appreciate your comments. Rectified unidentified abbreviations.   It is submitted that, with reference to past studies published in reputable journals, including the MDPI journal, the abstract presents key findings in the form of values/percentages (especially for research studies) to attract the attention of readers. However, if the respected reviewer still insists, we will remove values and percentages.   We would like to submit that we should mention the results, method, and conclusion in the abstract (for research studies), as per the Journal’s guidelines
2	The introduction section should usually not be subdivided. Please delete the subsection title 1-1	Thank you so much for your suggestion. We have deleted subsection 1.1, instead, a new section has been added.
3	Highlight the reminder of the manuscript in the last part of the introduction.	We appreciate your comments. The last paragraphs of the introduction reflect the justification, contribution, and research objectives (last 2 paragraphs in the Introduction).
4	I find that the introduction is unfocused and overly long, lacking a clear research question or motivation. Further, authors reiterate known information about the insurance scheme in Saudi Arabia without identifying gaps or clarifying how this study advances the field of health insurance.	We appreciate your comments. As suggested, we have identified the research gaps and research questions in the introduction section.
5	Page – 7. Results are overly descriptive and need to be interpreted with greater clarity. For example, How do you explain that participants living in urban areas are more likely to participate in the health insurance program as compared to those living in rural or suburban areas?	We appreciate the recommendations. We have tried to interpret the results and explained them with clarity, wherever required throughout the manuscript.
6	Page – 12. Line 364. What is Health Belief Model (HBM) and what is the relationship with the results above?	Thank you for the suggestion. We have deleted this paragraph as it was inappropriate in this context.
7	Subsections “Contributions, implications and limitations” should be integrated into the conclusion section.	Thank you for your suggestion. As per the Journal’s policy, we prefer to provide them separately, which would provide more clarity.   However, if the respected reviewer still insists, we will integrate them into the conclusion section

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The article “Factors Influencing Support for National Health Insurance: Evidence from Qassim Region, Saudi Arabia” is a well-structured research manuscript on a critical policy issue for Saudi Arabia. The study is relevant, the methodology is generally sound, and the findings are valuable.

With 1,194 participants, the study has strong statistical power, which is a significant advantage over many prior studies in the region. The use of contingent valuation is appropriate for a hypothetical product like NHI and the statistical analyses are well-chosen to answer the research questions. The results provide concrete data that can directly inform policy design.

The Abstract and Introduction section are overall well-written. The Introduction provides a solid background on Universal Health Coverage, Saudi healthcare challenges, and Vision 2030. The paragraph on Cooperative Health Insurance System is crucial but slightly long. Consider breaking it into two shorter paragraphs for readability. The justification for focusing on one region is excellent and clearly states the study's contribution.

Regarding the Materials and Methods section, the calculation starting from a population of 13.36 million is incorrect, as the study is focused only on the Qassim region. The sample size calculation should be based on the population of Qassim or justified as a national-level estimate. This needs clarification.

In the Results section, the narrative if the binomial regression analysis interpretation is good. The authors are encouraged to emphasize the key takeaways – being male, having a medium-sized family, and having a chronic disease increase the odds of WTP. Having private insurance, high income, and satisfaction with current services decrease the odds of WTP. Regarding Table 3, the authors could consider using ANOVA/t-tests to compare mean WTP amounts between groups, rather than relying on Chi-square.

The Discussion section effectively compares the findings of this study with previous studies in Saudi Arabia and internationally with a very comprehensive and honest Limitations sub-section.

Other than the mentioned, no further enhancements are needed.

Author Response

Reviewer -2
1	The paragraph on Cooperative Health Insurance System is crucial but slightly long. Consider breaking it into two shorter paragraphs for readability. .	Respected reviewer, we appreciate your comments. The paragraph has been shortened after removing unnecessary lines.
2	The justification for focusing on one region is excellent and clearly states the study's contribution	We are thankful for your comments. The contribution of the study is mentioned in the last paragraphs of the introduction section.
3	Regarding the Materials and Methods section, the calculation starting from a population of 13.36 million is incorrect, as the study is focused only on the Qassim region.The sample size calculation should be based on the population of Qassim or justified as a national-level estimate. This needs clarification.	We appreciate your comments. We regret the mistake in the figure. It is 1.336 million population (not 13.36 million) in the Qassim region. However, the sample size remains the same. .
4	In the Results section, the narrative of the binomial regression analysis interpretation is good. The authors are encouraged to emphasize the key takeaways – being male, having a medium-sized family, and having a chronic disease increase the odds of WTP. Having private insurance, high income, and satisfaction with current services decrease the odds of WTP. Regarding Table 3, the authors could consider using ANOVA/t-tests to compare mean WTP amounts between groups, rather than relying on Chi-square.	We highly appreciate your comments. We emphasized the key takeaways wherever applicable throughout the manuscript. Key takeaways are provided at the end of the discussion of the results (in tables), Table 3 has been modified using ANOVA/t-tests to compare mean WTP amounts between groups. The results of the ANOVA/t-tests are also modified.   The narrative of the binomial regression analysis has been modified according to the results.
5	The Discussion section effectively compares the findings of this study with previous studies in Saudi Arabia and internationally with a very comprehensive and honest Limitations sub-section.	We appreciate your comments. Thank you

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Authors,

Thank you for the opportunity to review your manuscript on Saudi citizens’ willingness to pay for a National Health Insurance (NHI) in Qassim. The topic is highly relevant to ongoing health sector reforms under Vision 2030 and the broader pursuit of sustainable healthcare financing. Your study collected valuable data from a large sample, and I appreciate the effort to analyze multiple angles of willingness (incidence of WTP and amount of WTP). However, in its current form, the manuscript requires major revisions before it can be considered for publication. Below, I provide detailed feedback and suggestions, which I hope will be constructive in strengthening your work.

1. Theoretical Framework and Literature Context:

Issue: The introduction sets the stage in terms of policy context (UHC, Vision 2030) but lacks a clear theoretical or conceptual framework regarding why and how we expect certain factors to influence WTP. This absence makes the study appear somewhat exploratory and descriptive, rather than driven by theory or hypotheses.

Recommendation: Introduce a theoretical perspective or conceptual model for WTP. Even if there isn’t a single established theory in this niche, you can draw from relevant economic or behavioral theories. For instance, theories of health insurance uptake might consider expected utility (people weigh cost vs expected benefit), or behavioral factors like perceived need and trust in the system. If socio-demographic factors are used as proxies, explain why each might matter (e.g., income affects ability to pay; education may correlate with awareness of insurance benefits; gender roles might influence risk perception or financial decision-making in the household; etc.). Providing this rationale up front will make your analytical choices clearer and your discussion of results more meaningful. Also, consider incorporating insights from broader literature. There is rich international research on willingness-to-pay for health insurance that could inform your approach.

Suggested additions: In the revised introduction, after discussing Vision 2030 and the rationale for NHI, add a paragraph summarizing key findings from previous WTP for health insurance studies globally and in the region. Highlight factors known to influence WTP and any gaps/inconsistencies. For instance: “Previous studies in other countries have observed that income and education tend to positively influence WTP for insurance, while age often has a negative effect. Household size can increase WTP, presumably due to greater family needs. However, factors like satisfaction with current services or chronic illness have shown mixed effects in different contexts.” This not only sets up your hypotheses but also positions your study within the scholarly conversation.

Also, consider a broader sustainability perspective. One crucial point often discussed in health financing literature is that relying on citizen contributions has limits—most systems still need substantial government subsidization.

2. Methodology—Sampling and Survey Design:

Issue (Sampling): Your method section reveals that you used a convenience online sample of Saudi adults in Qassim. This approach raises concerns about sample representativeness. Indeed, your achieved sample over-represents certain groups: urban residents (84% of sample), males (73%), and those with higher education (over two-thirds have a bachelor’s or higher, whereas the general population would have a much lower proportion of university graduates). This could introduce bias in the WTP estimates—for example, more educated individuals might be more aware of insurance benefits and thus more willing to pay (though your data showed mixed results on education, possibly due to confounding). The heavy skew towards male respondents is particularly concerning, as it may reflect the outreach method (perhaps the survey circulated in networks with more men). If women are underrepresented, yet appear more willing in percentage terms, the overall 49.3% willing figure might actually underestimate the true willingness in the population (since women’s voices were not equally captured). These biases limit the generalizability of the findings to the wider Qassim or Saudi population.

Recommendation: In the revision, be forthright about these sampling limitations. You have partially done so (your methods section notes that convenience sampling has limitations in generalizability), but this point should be emphasized in both the methods and the discussion of limitations. Explain why a convenience sample was necessary (e.g., logistical constraints, lack of sampling frame) but also acknowledge whom you might have missed (perhaps older adults not on social media, lower-income individuals without internet access, etc.). Importantly, consider if you can perform any post-stratification weighting or at least provide a table comparing your sample demographics to census demographics for Qassim. For example, if the population is ~50% female but your sample is 27% female, that is a discrepancy to report. While you may not be able to fully correct it, transparency is key. In the discussion, you should note that the WTP percentage and even some factor effects could be different if the sample were truly representative. For instance, since women (in your sample) had a higher willingness rate, a representative gender balance might yield a slightly higher overall WTP rate than 49%.

Issue (Survey design): The contingent valuation scenario needs more description. Currently, readers do not know exactly what respondents were told about the hypothetical NHI. This is crucial, because WTP can depend on scenario details. Were respondents, for example, told something like “imagine a National Health Insurance is implemented that covers XYZ services, would you be willing to pay for it monthly?”—and were they asked a yes/no first, then an amount? Or directly an open amount? Did you specify the payment mechanism (e.g. added payroll tax, or out-of-pocket premium)? Also, was “willingness to pay” framed per individual or per household? (You report it as per family per month—clarify if each respondent was answering for their household). All these details should be spelled out in the Methods section. If you used a bidding game or offered a list of amounts for the respondent to choose from, mention that. If it was open-ended, mention how you handled outliers (did some say extremely high amounts, did you cap it?). The credibility of your CV findings hinges on the design being sound and clearly communicated.

Recommendation: Expand Section 2.1 or add a subsection “Survey Instrument” to describe the NHI scenario and WTP elicitation format. Include the exact phrasing if possible. E.g., “Participants were given a description of the proposed NHI program (covering what benefits? perhaps mention if it’s supplementary to current services or replaces them) and then asked: ‘Would you be willing to pay monthly to join this program if it is implemented?’ Those who answered yes were further asked, ‘What is the maximum amount in SAR you would be willing to pay per month for NHI for you/your family?’.” If you did something differently, explain accordingly. Also, note any follow-up questions (some CV studies ask why a respondent is not willing to pay, etc.—did you include any such qualitative query? If yes, report it; if not, it’s fine).

Be mindful of potential biases in CV: for example, starting point bias (if you suggested a starting amount), strategic bias (people might under-report WTP expecting the government to pay more). You might not have data to assess these, but acknowledging them will show rigor. One suggestion: In the discussion, mention that because NHI is hypothetical, some respondents might be uncertain or distrustful, potentially affecting their stated willingness. Tying this to your finding about satisfaction (those satisfied with current services were less willing—perhaps they feel they don’t need to pay for improvements) could be insightful.

 

3. Results—Accuracy and Interpretation:

This is a crucial area needing attention. There are a few corrections and clarifications required:

• Correct misinterpretations: As noted, your text incorrectly interprets some results (gender, marital status, etc.). Please revise the results narration to align strictly with the data. For gender, clarify the difference between absolute numbers and proportions. It might be useful to report the percentage of each subgroup willing to pay: e.g., “67% of female respondents vs 43% of male respondents indicated willingness; however, because the sample had many more males, in absolute count more males were willing.” Then, when discussing the regression, you could say “controlling for other factors, being male was associated with higher odds of willingness (OR=2.789)—a finding that might initially seem counterintuitive given women’s higher raw percentage, but it likely reflects interactions with other variables (for instance, women in our sample tended to have slightly lower income or different age distribution, which the model adjusts for).” By doing this, you preempt confusion and show that you’ve thought about the result. If you are not sure why the regression result contradicts the descriptive pattern, simply acknowledge it as an area for further exploration rather than ignoring it.
• Marital status: Your chi-square showed no significant effect, so your text should not claim one. Remove or correct that statement. If marital status truly had no impact, you can just say “There was no statistically significant difference in WTP between married and unmarried respondents (approximately half of each group were willing to pay).” This is still a finding—sometimes null results are as important to report to dispel assumptions (maybe one might expect unmarried individuals, having fewer dependents, to be less willing, but you found no difference).
• Education: The way results are described for education is a bit confusing. Chi-square found an overall association (p<0.05), but your breakdown didn’t indicate a monotonic trend (PhD category had fewer willing proportionally than Bachelor’s, etc.). Then your linear regression found education not significant for amount. It’s okay that these are complex; just describe them clearly. Possibly bachelor’s holders formed the largest group and had moderate willingness. You might note: “Respondents with a Bachelor’s degree formed the plurality of our sample and about 48% of them were willing to pay, similar to the overall average. Those with only high school had a slightly higher willingness rate (~57%), while those with PhDs had a lower rate (~42%), though the sample sizes for the extremes were smaller. Overall, education level showed a modest association with willingness (chi-square p<0.05) but with no clear linear trend. In regression analysis, education did not emerge as a significant predictor of the amount willing to pay.” This level of detail shows you’ve parsed the data carefully.
• Income effect: You found that the highest willingness (in terms of amount) was in an upper-middle income bracket (18,000–24,000 SAR/month) rather than strictly increasing with income—and that those in 25k–35k had lower odds to join NHI (OR ~0.49). This is interesting and should be discussed. Is it a statistical quirk or perhaps wealthier people feel they can pay out-of-pocket or already have private insurance? You partially explain that those with private insurance were less likely to join NHI (which makes sense—they might not want to pay for something they feel they don’t need). Maybe many in the 25k–35k bracket have private insurance through work, hence their lower willingness. If so, that interaction can be mentioned: “Interestingly, the odds of willingness were lower in upper-middle income households (25–35k SAR) than in the lowest income group, according to the logistic regression. One possible explanation is that higher-income Saudis often already have private insurance benefits (indeed, our data show higher income correlated with having insurance coverage), so their incentive to enroll in NHI is less.” Including such reasoning will show thoughtful interpretation.
• Satisfaction and obstacles: You have intriguing findings here: people satisfied with current services were less willing to pay (OR ~0.64), and those facing obstacles to access were also less willing (OR ~0.54 for willingness, but then a positive beta for amount in linear regression). This seems counterintuitive at first—one might expect dissatisfied people to not trust the system and hence not want to pay, but your result is opposite: dissatisfied (not satisfied) folks were more willing. Actually, that makes sense: if you’re unhappy with the free public system, you might support an NHI reform hoping it improves things, whereas if you’re content, you don’t see a need to pay. Emphasize this logic in discussion. The obstacles result is a bit puzzling: those with no access problems were more willing to join NHI (perhaps because they are generally better off or have had good experiences and thus trust the idea?). Yet your linear regression suggests those who do face obstacles are willing to pay 10% more in amount. How to reconcile: it could be that among the subset who are willing at all, those with obstacles have higher willingness to pay (maybe they are really desperate for improvement, so they’d pay a lot if they agreed to pay), but a significant portion of them might have said “no” outright (maybe due to mistrust or inability), reducing their overall participation odds. This is a subtle point—you may or may not include it, but if you do, explain it carefully. At minimum, clarify in the results that “lack of obstacles was associated with higher likelihood of willingness (chi-square and logistic findings), whereas among those willing, people who had faced obstacles tended to state somewhat higher payment amounts (as per the regression on amount).” Being explicit will prevent readers from thinking you have contradictory findings—instead they’ll see it as a nuanced situation.
• Data completeness: You mention you excluded 12 incomplete responses (1206 -> 1194). That’s fine. You might want to mention if there were any extreme outliers in the WTP amounts (e.g., did someone write 10,000 SAR? If so, did you treat it as valid or consider it a protest/unrealistic bid?). If you have a maximum value, reporting the range might be useful (min, max of stated WTP). Right now, we only know median=100, mean=158, SD ~118 SAR. It sounds like most values were reasonable. If there were any cases of unrealistically high WTP, you could mention how you handled them (perhaps none were removed, which is okay).
•  

4. Discussion and Contextualization:

Beyond correcting the coherence issues already noted, the discussion should be enriched in the following ways:

• Integrate with literature: When you compare your findings to Riyadh/Jeddah studies that found ~66% willingness, provide some speculation or reasoning. For example: “Our WTP rate (49%) is lower than the ~66% found in prior surveys in Riyadh and Jeddah【12,29,31,32】. This difference could stem from regional variations—Qassim is less urbanized and on average has lower income levels than Riyadh/Jeddah, which might reduce willingness. It could also be due to differing sample compositions or the timing and framing of the surveys. Notably, our study was conducted in late 2024 when economic conditions (post-COVID, etc.) might have affected willingness.” Any plausible explanation is better than a simple statement of contradiction. Also, if any national or other regional studies exist (perhaps you cited [13-16] about expanding insurance and OOPE reductions—if those include WTP or related evidence, draw them in).
• Sustainability and policy implications: Since the journal (and the framing) is about sustainability, discuss what your findings imply for the sustainable implementation of NHI. For instance, you found the average household is willing to pay ~158 SAR/month. Is that enough to fund the system? Probably not fully—perhaps reference what portion of current health spending that would cover if, say, half of households paid that. If you have any estimate (even rough): e.g., 158 SAR/month per family is ~1896 SAR/year; if half of, say, 2 million families in Saudi paid that, it yields X billion SAR, which is some percentage of total health expenditure. This kind of insight could highlight that while helpful, NHI contributions would complement, not replace, government funding. This ties back to the literature suggestion that household premiums can’t be the sole financing.
• Broader benefits of citizen input: You mention public opinion being crucial for successful reforms (and cite sources for that). You could elaborate that finding out WTP is part of gauging public readiness and designing premiums that are acceptable. Here incorporate a citation like Ghosh & Ray (2012). One key takeaway from that article is the need for frameworks to manage complex social changes with stakeholder involvement. Draw an analogy like: “Reforming health financing is a complex, multi-stakeholder endeavor (government, public, private insurers). It requires understanding the “multidimensional” nature of public attitudes and behavior in policy change. Our study contributes one dimension (financial willingness), but other dimensions (cultural acceptance, trust) must also be addressed.” By doing so, you acknowledge that WTP is not the only factor in implementing NHI, which lends a realistic tone to implications.

• Policy recommendations: In your conclusion, you could add a sentence or two on what the government or policymakers should do in light of your findings. For example: “Policymakers should note that willingness to pay is far from universal; extensive public awareness campaigns and education about NHI benefits may be necessary to increase buy-in, especially among groups currently less willing (e.g., those with lower perceived need or those who are older). Also, the contribution levels citizens are willing to pay (median ~100 SAR) suggest that premiums must be set affordably to maximize participation—a balance must be struck between financial sustainability and enrollment.” This kind of recommendation flows from your data and emphasizes sustainability (keeping premiums affordable to ensure a broad risk pool, etc.). If any of the references above discuss premium affordability or related policy tactics, you could tie that in.

 

5. Minor Edits and Presentation:

• Abstract: Once you revise the main text, ensure the abstract is updated to reflect the changes accurately. The abstract currently has that p<0.001 issue which should be removed or clarified. Also, mention the key significant factors in the abstract results (now you list some, but make sure it matches the final analysis outcome after any corrections).
• Keywords: Consider adding “Universal Health Coverage”, “Contingent valuation”, and “Willingness-to-pay” as keywords if they aren’t there. These will help in indexing and were central to the study.
• Tables/Figures: Double-check table numbering and titles after revisions. Ensure all acronyms are explained in a footnote or note (e.g., SAR = Saudi Riyal, WTP = willingness to pay, etc., on first appearance). If you decide to add a figure of WTP distribution, label it properly. Also, correct any inconsistencies (for example, Table 5’s caption was cut off in the PDF I saw—make sure it reads fully “...on amount willing to pay”).
• Proofreading: As noted in the language section, do a thorough proofreading. It might help to read the manuscript aloud or have a colleague with strong English skills review it. Pay attention to subject-verb agreement, proper tense (e.g., use past tense for describing what you did in the study, present tense when stating what the paper “shows” or what others “report” generally), and eliminate any redundant words.

In conclusion, your study has valuable data and can make a meaningful contribution, but it requires significant revision to reach its potential. The effort will be well spent: clarifying your arguments and situating them in the larger literature will greatly enhance the impact of your work. I recommend “Reconsideration after Major Revisions”, and I encourage you to systematically address each of the points above. Good luck with your revisions—I am confident that if you implement these changes, the manuscript will be much improved.

Thank you for your attention to these comments. I look forward to seeing a revised version of your work.

Comments on the Quality of English Language

The manuscript’s English is mostly clear and the meaning can be followed, but there are numerous minor errors and some awkward phrasing that should be addressed to ensure a professional presentation. Examples of language issues include:

• Grammar and Word Choice: There are instances of incorrect preposition or verb usage (e.g., “contradicts with findings of the studies...” should be “contradicts the findings of other studies”; “scored high for being willing to pay” is a colloquial and unclear way to say “had a higher willingness to pay”). The text sometimes uses unnecessary hyphenation (perhaps due to justification in PDF, like “willing-ness” and “health-re-lated” in the abstract—these appear to be line-break artifacts and should be removed). Ensure words are correctly joined (e.g., “healthcare” vs “health care” used consistently).
• Sentence structure: A few sentences are overly long or convoluted. For example: “Most research on this topic has focused on major economic hubs... with limited studies covering national samples, and those that do often have small sample sizes [12,29-35].” This is understandable but could be split for clarity. In the results and discussion, some sentences listing multiple statistics could be broken up to improve readability.
• Tone and academic style: The tone is mostly appropriate, but there are places where more formal academic wording is needed. For instance, “scored high on not having WTP” is not a standard way to describe a result—something like “a higher proportion were not willing to pay” would be clearer. Also, avoid subjective words like “most importantly” when describing results (the data should speak for itself as to what is important, or the importance should be justified, not just asserted).
• Consistency: The paper should maintain consistent terminology. Sometimes “willingness to pay” is written in full, other times abbreviated as WTP—that’s fine if defined, but ensure the abbreviation is introduced early (it is in the abstract) and consistently used thereafter. Also, the NHI scheme is sometimes referred to as “health insurance program,” “health insurance scheme,” etc.—that’s okay, but consider capitalizing NHI throughout once defined, to keep it clear that it refers to the specific proposed program.
• Punctuation: There are minor punctuation issues (missing commas in some long sentences, etc.). Also, check the formatting of percentages and p-values (e.g., “p=0.051” should ideally be written with a leading zero and perhaps two decimal places if not significant, i.e. p = 0.051, and consistently use either p < 0.05 or p = 0.051 notation).

Overall, the language is at a level that one can comprehend the manuscript, but it would benefit from careful proofreading by a native or proficient English speaker, or an academic editor. Polishing the language will also help ensure that the substantive issues (like those discussed above) are communicated clearly without ambiguity. Given the number of statistical findings being reported, clarity in language is essential to avoid misunderstandings.

Author Response

Dear Reviewer and Academic Editor 

Reviewer 3 has given detailed comments which are all addressed in red color in main manuscript, however, due to words limit restriction it is not possible to copy and paste here, but the file in which detailed response is given is attached title "response to Reviewer-3. 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I'm satisfied with the author's responses, finding the revised manuscript easier to follow, and recommends acceptance. 

Author Response

Dear Sir, 

We are grateful to you for your efforts and for recommending our manuscript for publication. Thank you 

Reviewer 3 Report

Comments and Suggestions for Authors

The submitted manuscript tackles a timely policy problem for Saudi Arabia by assessing citizens’ willingness to financially support a proposed National Health Insurance (NHI) program. Using a contingent valuation (CV) survey of heads of household in the Qassim region, the authors report that about half of the sample supports NHI and would pay a mean monthly contribution of 158 SAR, with significant predictors including gender, household size, chronic disease, and satisfaction with current services. The paper is ambitious and could make a useful contribution, but substantial revisions are needed to sharpen the narrative, strengthen the methodology, and situate the work within broader scholarship.

 

1. Contextualization and theoretical framing.

The introduction provides a broad overview of Saudi healthcare reforms but does not sufficiently connect the study to theories of social insurance, public goods and industrial relations. Much of the background repeats basic statistics about demographics and costs, yet it seldom discusses why willingness‑to‑pay (WTP) varies across contexts or how public trust and stakeholder relations shape insurance uptake. To ground the work in a richer theoretical frame, the authors should integrate insights from industrial relations and policy design literature. For example, Ghosh & Ray’s article on contemporary models for industrial relations offers a global perspective on aligning employer, employee and government interests in social welfare programs [Ghosh, A., & Ray, P. (2012). A contemporary model for industrial relations: Relook from global perspective. Management and Labour Studies, 37(1), 17-30]. Their discussion of participatory policy‑making and labour‑management cooperation is directly relevant because successful NHI implementation will require negotiated contributions from workers and employers; citing and engaging with this work would help the authors explain why stakeholder voice and institutional trust are critical components of willingness to support NHI.

Similarly, recent contingent valuation studies outside Saudi Arabia should be discussed. A systematic review and meta‑analysis of 30 studies on community‑based health insurance in Ethiopia [Kaso, A. W., Debela, B. G., Hareru, H. E., Ewune, H. A., Debisa, M. A., Sisay, D., & Hailu, A. (2025). Willingness to join community-based health insurance and associated factors among households in Ethiopian: a systematic review and meta-analysis. Cost Effectiveness and Resource Allocation, 23(1), 12] synthesized evidence from cross‑sectional surveys and found that only about 60 % of households were willing to join such schemes; determinants included age, education, wealth status, knowledge/awareness and chronic illness, and the authors concluded that trust, affordability and perceived quality of care strongly influenced willingness.

Another recent study from rural Senegal [Sokhna, C., Boyer, S., & Ventelou, B. (2025). Uptake of and willingness to pay for health insurance in rural Senegal: a reinforcement effect. BMJ Public Health, 3(1)] analysed uptake and willingness to pay using a simultaneous‑equation model and reported that only one‑third of households were aware of community‑based insurance and that the mean willingness to pay (≈3,865 CFA francs) exceeded the premium; importantly, being enrolled in the scheme increased the maximum amount individuals were willing to pay by about 41%. Comparing such cross‑country evidence would allow the authors to explain whether the relatively low WTP observed in Qassim reflects sampling differences, socio‑economic context or methodological factors, and would demonstrate engagement with global research.

 

2. Research design and sampling.

The paper states that a cross‑sectional CV survey was administered via an online platform and that convenience sampling was employed. Although 1,194 completed questionnaires are reported, the sampling strategy undermines external validity. Convenience sampling through social media likely excludes older individuals, low‑income households without internet access, and residents in rural areas. The authors should either redesign the sampling to be probability‑based (e.g., multistage cluster or stratified random sampling) or, at minimum, acknowledge more clearly that the findings cannot be generalized to the entire Saudi population.

Recent international work illustrates how rigorous sampling and modelling can address these issues: the Senegal study employed stratified random sampling and a Heckman‑type selection model to account for awareness and enrolment biases, while the Ethiopian meta‑analysis highlighted that studies using community‑based insurance often oversample informed and wealthy households and advocated for more representative designs. Another methodological concern is that only heads of household were surveyed; this implicitly assumes that intra‑household preferences are homogeneous and may introduce gender bias. The paper would be strengthened by including multiple adult members per household or, at least, by discussing how decision‑making dynamics might affect WTP.

 

3. Measurement and analysis.

The contingent valuation scenario is described, but the choice of a payment‑card elicitation with mid‑point values warrants further justification. CV methods are prone to hypothetical bias, anchoring effects and strategic behaviour, particularly when respondents are unfamiliar with insurance products. The authors should explain how they minimized such biases—e.g., by piloting bids, using follow‑up certainty scales or employing ex‑ante cheap talk. Also, the decision to record amounts greater than 500 SAR as 525 SAR could distort the distribution of WTP. Instead, an open‑ended question or a two‑step dichotomous choice design could provide cleaner data. The analytic strategy relies on binomial logistic regression, dichotomizing WTP into willingness vs. unwillingness. This approach discards information contained in the magnitude of contributions and contributes to low explanatory power (Nagelkerke R² of 0.147). More informative modelling techniques—such as ordered probit, Tobit or generalized linear models—could exploit the continuous WTP data and better capture heterogeneity. Furthermore, the models omit potentially important determinants such as perceptions of healthcare quality, trust in government, risk attitudes and employment status. Including such variables would likely improve model fit and generate more nuanced insights.

 

4. Results presentation and interpretation

The results section contains descriptive statistics and regression outputs, but the narrative is sometimes confusing. For instance, the authors state that being male increases the odds of willingness (OR = 2.79), yet the multiple regression shows a negative coefficient for gender, implying that men are willing to pay less. These contradictory findings suggest either coding errors or misinterpretation; the variables and reference categories should be clearly defined, and results should be internally consistent. The tables are poorly formatted in the PDF and include typographical errors; they should be edited for clarity. It would also help to report goodness‑of‑fit measures, test statistics and confidence intervals uniformly. The discussion of regression results sometimes overstates relationships that are statistically insignificant; for example, the authors infer that urban residents are less willing to pay despite the p‑value of 0.031 being marginal and the R² very low. The authors should avoid causal language and acknowledge the modest explanatory power of their models.

 

5. Discussion and integration with literature.

The discussion appropriately compares mean WTP estimates with prior studies in Riyadh and Jeddah but does not critically examine why the Qassim region differs. It would benefit from a deeper engagement with recent international literature on WTP for social health insurance. For example, the Ethiopian meta‑analysis reported a pooled willingness to join of 60.42 % and identified determinants such as education, wealth, knowledge/awareness, health status and trust. This finding contrasts with the present study’s higher mean WTP but lower participation rate and suggests that awareness and perceived quality of care are critical drivers of demand. The Senegal study further demonstrated a reinforcement effect, with enrolment raising willingness to pay by roughly 41 %; its authors noted that only 33 % of households knew of community‑based insurance, underscoring the importance of information campaigns. Such contrasts could prompt a richer exploration of cultural values, public trust and economic incentives. The authors should also cite systematic reviews of contingent valuation methods for healthcare, which discuss common biases and best practices, and should reference Ghosh & Ray’s framework to underscore the need for inclusive policy dialogue during NHI implementation. Doing so would demonstrate awareness of interdisciplinary scholarship and inform more balanced policy recommendations.

 

6. Conclusions and policy implications

The conclusion asserts that NHI is both feasible and socially justified and suggests that the government can rely on citizen contributions to supplement healthcare financing. These claims are stronger than warranted by the data. Because WTP was measured under a hypothetical scenario and the sample is not representative, the estimated mean contribution should not be extrapolated to national budgets. Moreover, the modest model fit and contradictory coefficients indicate that the determinants of WTP are still poorly understood. The conclusion should be tempered to emphasize that the findings are preliminary and regional, and that further nationally representative research is needed. Policy recommendations should also consider the administrative and equity challenges of introducing NHI, including premium affordability for low‑income households, employer participation and the regulatory framework—topics that the industrial relations literature addresses.

Author Response

Respected Sir, 

We are highly thankful for your precious comments/suggestions to refine our manuscript. We have submitted the revised manuscript along with a file on responses from authors based on your detailed comments.

Thanking you 

Author Response File: Author Response.pdf

Round 3

Reviewer 3 Report

Comments and Suggestions for Authors

.