Insurance coverage and treatment gaps among adults with hypertension or diabetes in Kenya: a sex-stratified analysis of the 2022 Demographic and Health Survey

A sex-stratified analysis of the 2022 Kenya Demographic and Health Survey reveals that despite nearly half of diagnosed adults with hypertension or diabetes having health insurance, substantial treatment gaps persist across all wealth and sex groups, suggesting that expanding insurance enrollment alone is insufficient to improve chronic care access without addressing deeper systemic barriers like medicine availability and affordability.

The Gist

Imagine Kenya's healthcare system as a massive, complex bus network designed to help people who have been diagnosed with two specific, long-term conditions: high blood pressure (hypertension) and diabetes.

In October 2024, the government decided to replace the old bus company (the National Health Insurance Fund, or NHIF) with a brand new one called the Social Health Authority (SHA). Before the new company started running the routes, researchers took a snapshot of the old system using data from 2022. They wanted to see: Who had a ticket (insurance)? Who was actually riding the bus (getting treatment)? And who was left standing at the stop?

Here is what they found, explained simply:

1. The "Ticket" Problem: Not Everyone Has a Pass

Even among people who knew they were sick, less than half had a ticket to ride the bus.

• The Stat: Only about 47% of diagnosed adults had any kind of health insurance.
• The Analogy: Imagine a stadium where half the fans are told they need a ticket to get in, but nearly half of them are still standing outside the gates.
• The Inequality: The tickets were mostly held by the wealthy. If you were rich, you were much more likely to have a ticket. If you were poor, you were likely standing outside.

2. The "Bus Ride" Problem: Having a Ticket Doesn't Mean You Get on

This is the most surprising part of the study. Even for the people who did have a ticket, the bus often didn't show up, or they couldn't get on it.

• The Stat: About 64% of people who knew they were sick were not taking their daily medication.
• The Analogy: Imagine you have a valid bus pass, but when you get to the stop, the bus is empty of seats, the driver is missing, or the bus is out of gas. You have the permission to ride, but you still can't get to your destination.
• The Result: It didn't matter if you were rich or poor; the "bus ride" (treatment) was missing for everyone. Even among the richest people, about 60% weren't getting treated.

3. The Gender Gap: A Harder Climb for Women

The study looked at men and women separately and found a distinct difference in their struggles.

• Men: Had slightly better access to tickets (insurance) than women.
• Women: Faced much bigger hurdles. When women didn't have insurance, they reported that getting money for treatment and traveling to the clinic were "big problems" far more often than men did.
• The Analogy: If the bus stop is a mountain, men were at the bottom with a slightly better map. Women were often further up the mountain, and without a ticket, they found the climb nearly impossible because they couldn't afford the gear (money) or the path was too far away (distance).

4. The Big Takeaway: Buying a Ticket Isn't Enough

The researchers concluded that simply handing out more tickets (enrolling more people in insurance) won't fix the problem if the buses aren't running.

• The Reality: The study found that having insurance didn't strongly guarantee that a person was actually taking their medicine.
• The Lesson: It's like having a gym membership but the gym is closed, or the equipment is broken. To fix the health crisis, Kenya needs to do more than just sign people up for insurance. They need to make sure the "buses" (medicines and clinics) are actually there, working, and ready to pick people up.

Summary

Before the new health authority took over, Kenya faced a double challenge:

1. Incomplete Coverage: Too many people didn't have insurance.
2. Broken Service: Even those with insurance often couldn't get the treatment they needed.

The study serves as a "before picture" for the new system. It warns that if the new system only focuses on getting more people to buy tickets without fixing the broken buses (medicine availability and clinic readiness), the treatment gaps will remain wide open.

Technical Summary

Technical Summary: Insurance Coverage and Treatment Gaps Among Adults with Hypertension or Diabetes in Kenya

Problem Statement
Kenya is facing a significant burden of non-communicable diseases (NCDs), specifically hypertension and diabetes, which drive morbidity, premature mortality, and household financial stress. While previous literature has established low insurance coverage and poor treatment cascades in the general population, there is a critical evidence gap regarding adults who have already been diagnosed with these conditions. Specifically, it remains unclear to what extent these diagnosed individuals were financially protected and maintained on treatment at the close of the National Health Insurance Fund (NHIF) era, just prior to Kenya's transition to the Social Health Authority (SHA) in October 2024. Without a nationally representative pre-transition baseline, it is difficult to assess the efficacy of the new SHA reforms in closing treatment gaps.

Methodology
This study is a cross-sectional secondary analysis of the 2022 Kenya Demographic and Health Survey (KDHS 2022), the last nationally representative survey conducted before the SHA transition.

• Data Source: The analysis utilized de-identified microdata from KDHS 2022, merging household roster insurance data with individual interview files.
• Study Population: The analytic sample comprised 1,932 adults (1,384 women aged 18–49 and 548 men aged 18–54) who reported a prior clinician diagnosis of hypertension and/or diabetes.
• Definitions:
  • Treatment Gap: Defined as a respondent with a diagnosed condition who was not currently taking medication for at least one of their diagnosed conditions ("diagnosed-not-on-treatment").
  • Insurance Exposure: Categorized as "any insurance" and specifically "NHIF coverage," derived from household roster questions.
• Statistical Analysis: Analyses were survey-weighted and sex-stratified to account for the complex sampling design. The study employed:
  • Weighted prevalence estimates with 95% confidence intervals.
  • Concentration indices to quantify wealth-related inequality in insurance coverage and treatment gaps.
  • Survey-weighted generalized linear models (quasi-Poisson family with log link) to estimate adjusted prevalence ratios (APR) for the association between insurance status and treatment gaps, controlling for age, wealth, education, residence, employment, and diagnosis profile.
  • Sensitivity analyses using a stricter "no-treatment" definition and Erreygers-corrected concentration indices.

Key Results

• Insurance Coverage: Overall, 47.7% of diagnosed adults had any health insurance, with NHIF being the dominant provider (43.4% of the total sample). Coverage was significantly higher among men (53.7%) than women (42.7%).
• Treatment Gaps: A substantial treatment gap existed, with 63.8% of diagnosed adults reporting they were not currently on medication for their condition. This gap was higher in women (67.1%) than men (59.8%).
• Inequality: Insurance coverage was strongly "pro-rich," indicated by a positive concentration index (0.2889). In contrast, treatment gaps were distributed across the wealth gradient, with a concentration index near zero (−0.0347), indicating that lack of treatment affected all wealth quintiles, with gaps exceeding 60% in every quintile.
• Association between Insurance and Treatment: After adjusting for covariates, insurance coverage was not strongly associated with a lower prevalence of treatment gaps. The adjusted prevalence ratio (APR) for insured versus uninsured adults was 0.91 for women and 0.93 for men, neither of which was statistically significant.
• Access Barriers: Among women, the uninsured reported significantly higher barriers to care compared to the insured. Specifically, 63.2% of uninsured women cited "money for treatment" as a big problem, compared to 30.6% of insured women. Similar disparities were observed regarding distance to facilities.
• Diagnosis Profile: Treatment gaps were highest among those with hypertension only (69.1% pooled) and lowest among those with both hypertension and diabetes (37.9% pooled), though the latter subgroup had higher insurance coverage.

Key Contributions

1. Baseline Establishment: The study provides the first nationally representative pre-transition baseline for insurance coverage and treatment continuity among diagnosed hypertensive and diabetic adults in Kenya, offering a critical benchmark for evaluating the new Social Health Authority (SHA) reforms.
2. Decoupling of Insurance and Treatment: The analysis challenges the assumption that insurance enrollment automatically translates to effective treatment. It demonstrates that even among those with insurance, treatment gaps remain substantial, suggesting that financial coverage alone is insufficient to ensure medication adherence or access.
3. Inequality Nuance: The paper distinguishes between inequality in access to insurance (which is wealth-driven) and inequality in treatment gaps (which are widespread across all wealth groups). This suggests that the barrier to treatment is systemic and affects the poor and non-poor alike once a diagnosis is made.
4. Gender-Specific Insights: The study highlights that while men have higher insurance coverage, women face larger treatment gaps and significantly higher reported financial and geographic barriers when uninsured.

Significance and Claims
The authors position this work as a critical policy intervention point. They argue that the findings indicate Kenya faced two interlinked problems at the end of the NHIF era: incomplete enrollment in a fragmented insurance landscape and, more critically, incomplete effective coverage even among the enrolled.

The paper claims that expanding insurance enrollment alone is insufficient to close treatment gaps. Instead, the authors suggest that the SHA transition must be accompanied by improvements in:

• Benefit depth: Ensuring insurance packages cover necessary chronic care medicines.
• Medicine availability: Addressing stock-outs and supply chain issues.
• Frontline readiness: Strengthening the capacity of facilities to deliver chronic care.
• Gendered barriers: Specifically addressing the financial and distance constraints that disproportionately affect women.

The study concludes that without these systemic improvements, the magnitude of treatment gaps documented in 2022 is likely to persist under the new SHA framework, regardless of enrollment targets. The authors explicitly state that their cross-sectional design precludes causal inference and that the findings reflect a snapshot of the system's performance at a specific historical moment.