I would agree with you, but if you dig into the details, the original paper was written in March 2015. This is the retracted article here: https://pubmed.ncbi.nlm.nih.gov/25722125/ Note the "Retracted article" text.

They said the data was inaccurate, so they would republish the study with corrected data.

That link above says "Corrected and republished" in the Sept 2015 edition. OP's link is the corrected article. Note the September 2015 date: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4857885/

The Deseret News article says "The study, published in the Journal of Health and Social Behavior, and covered by the Deseret News, among other media outlets, found a 6 percent higher likelihood of divorce for couples with an ill wife."

The updated research paper:

Neither husband’s nor wife’s cancer or lung disease onset is associated with subsequent divorce. Wife’s heart problems onset (but not husband’s) is positively associated with a two percent higher probability of subsequent divorce compared with staying married. Further, we reject a null hypothesis of the equality of the heart problem onset coefficients for husband and wife at the p<0.05 level, providing robust support that the relationship between heart problem onset and divorce is gendered. Wife’s stroke onset is positively associated with a three percent higher probability of divorce compared with staying married, however this association is not statistically significantly different from the coefficient for husband’s stroke (p= 0.3818), which is itself not statistically significant.

The updated research paper says that cancer and lung disease was NOT associated with increased divorce rate for either husbands or wives. Heart disease in women was associated with a 2% higher risk of divorce. Stroke in women also showed a 3% increase of divorce, but the sample size was so small that it wasn't considered statistically significant.

So, the retracted study claimed a 6% increased divorce. The updated paper found a 2% increase for women with heart problems. No increased divorce for other health problems. A much smaller result.

Across the four major health problems, three of them were not associated with higher divorce among sick men or sick women. It's odd that ONLY heart disease in women was associated with an increased chance of divorce - and only by 2%. It might just be a fluke finding that means nothing. "Statistical significance" is a mathematical way of evaluating data, and it basically says, "we only count something as statistically significant if it has a 5% chance of happening by random chance". What that means is that you can get finding that are "statistically significant" when they're actually just noise in the data.

Some nerdy comments about "statistical significance": let's say that you said you were going to use telekinesis to make a coin land on "heads". You then flipped a coin 5 times and it happened to land on heads every time. The odds of that happening are about 3%. That result would be considered "statistically significant". You could then write an article that says "we found evidence that telekinesis exists and that you can cause a coin to land on heads". But in reality: maybe it was just a fluke. Also, when you're evaluating four different illnesses and outcomes for both men and women, then there are eight possible outcomes. That increases the odds of a statistically significant result by 8-fold.